Search

In 2016, repeated post-exercise head-out 40°C water immersion emerged as a practical heat acclimation strategy that improves endurance exercise performance in the heat in physically active individuals. However, it remains to be determined whether this heat acclimation strategy could improve endurance performance in the heat in well-trained endurance athletes. It would also be of interest to examine inflammatory and oxidative stress responses to this heat acclimation strategy. Addressing that question is relevant, given that the degree of hyperthermia associated with postexercise 40°C water immersion induces inflammation and oxidative stress, and knowing that both physiological events can impact exercise performance. Another concern related to the post-exercise 40°C water immersion intervention is the risk of heat illness. Therefore, it is important to identify a thermometric method that would be suitable for monitoring body core temperature during hot water immersion sessions in ‘real life’. Study One examined the effects of repeated post-exercise head-out hot (40°C) water immersion on physiological and exercise performance outcomes in warm and hot environmental conditions in a group of well-trained non-heat acclimated male endurance athletes. Methods: Sixteen well-trained male cyclists completed a six-day intervention involving a daily cycling exercise for 40 minutes at 50% of their peak power output in a cool environment (14°C, 40% RH) followed immediately by either hot water immersion (HWI; n = 8) or thermoneutral water immersion (CON; n = 8) for 40 minutes. At baseline and post-intervention, participants completed two 30-minute continuous cycling tests followed by 20-km time trial tests (one at 27°C, 40% RH and the other one at 35°C, 40% RH) on two different days separated by one day of rest. Results: The HWI group showed a significantly decreased peak heart rate (-7.37 ± 5.21 beats∙min-1, p = 0.03), peak thermal sensation (-0.56 ± 0.41 arbitrary units, p < 0.01), and rating of perceived exertion (-1.00 ± 0.75 arbitrary units, p = 0.02) during the 30-minute continuous test at 27°C. The HWI group also showed a significantly decreased peak thermal sensation (-0.50 ± 0.53 arbitrary units, p = 0.03) and peak rating of perceived exertion (-1.62 ± 1.06 arbitrary units, p = 0.01) during the 30-minute continuous test at 35°C. None of these variables were altered in the CON group. Furthermore, the HWI group showed only a tendency for improved 20 km time trial performance at 27°C (p = 0.06) and 35°C (p = 0.06). Conclusion: The findings from this study indicate that a six-day post-exercise hot water immersion reduces cardiovascular and perceptual strain during exercise at 27°C and reduces only perceptual strain during moderate-intensity exercise at 35°C. Although the postexercise hot water immersion intervention did not significantly improve the 20 km time trial performance either at 27°C or 35°C, the reduction in the completion time of the latter test by 1.77% in the HWI group can be considered practically significant. Study Two determined the acute and chronic effects of exposure to post-exercise head-out immersion in 40°C water on plasma cytokine response, oxidative stress, and antioxidant capacity. Methods: Fourteen well-trained male cyclists completed a sixday intervention involving a daily cycling exercise for 40 minutes at 50% of their peak power output in a cool environment (14°C, 40% RH) followed immediately by either hot water immersion (HWI; n = 7) or thermoneutral water immersion (CON; n = 7) for 40 minutes. Ten ml of a venous blood sample was taken before and after the first session to assess acute inflammatory responses, oxidative stress, and total antioxidant capacity. Resting venous blood was also sampled 48 hours after the last session to assess chronic inflammatory responses, oxidative stress, and total antioxidant capacity. The inflammatory response was assessed by measuring the following cytokines: tumour necrosis factor (TNF)-α, interleukin (IL)-1β, IL-1 receptor antagonist (IL-1ra), IL-6, and IL-10. Oxidative stress was determined by measuring thiobarbituric acid-reactive substances (TBARS), whereas total antioxidant capacity (TAC) was measured as a sum of all antioxidants in the plasma. Results: Acute postexercise hot water immersion had no significant effects on any of the measured biomarkers (all p > 0.05). Acute post-exercise immersion in 34°C water significantly increased IL-6 (p < 0.01). Post-exercise hot water immersion over six consecutive days significantly increased resting plasma IL-1β concentration (p < 0.01). Conclusion: This study suggests that chronic heat stress imposed by post-exercise hot water immersion induces an inflammatory response but not oxidative stress in welltrained male endurance athletes. Study Three examined whether commercially available tympanic thermometers could be a suitable substitute for the expensive ingestible telemetric pills or invasive rectal probes for monitoring core temperature during the post-exercise head-out 40°C water immersion. Methods: Sixteen male cyclists cycled for 40 minutes at 50% of their peak power output in a cool environment (14°C, 40% RH), after which they were immersed in a bath of hot water for 40 minutes. Participants' tympanic and core temperatures were measured at rest, after exercise, and every 10 minutes throughout the hot water immersion session. The tympanic temperature was measured by GeniusTM 2 and Braun Pro 4000 Thermoscan tympanic thermometers, whereas core temperature was measured via an ingestible telemetric pill. The latter was used as a reference against which the tympanic thermometers were validated. Results: No statistically significant differences in temperature readings were observed between the telemetric pill and Braun Pro 4000 Thermoscan at any time point during the hot water immersion session, and these temperature readings were significantly correlated at all time points. The overall bias in temperature reading provided by Braun Pro 4000 Thermoscan relative to the telemetric pill was within the acceptable limit (< 0.3°C). Temperatures provided by GeniusTM2 at the 20-, 30, and 40-minute time points during the hot water immersion period were significantly higher than those of the telemetric pill, and there were no correlations between the devices at the last two time points. The overall mean bias associated with the GeniusTM2 tympanic thermometer was 0.50°C. Conclusion: The findings from this study indicate that Braun Pro 4000 Thermoscan could be a suitable tool for monitoring core temperature during post-exercise head-out 40°C water immersion. Unfortunately, GeniusTM2 did not pass the validity testing and thus it is not recommended for use during hot water immersion.

Autophagy/mitophagy are cellular processes that play pivotal roles in maintaining mitochondrial function; thus, understanding their responses to high-intensity interval exercise (HIIE) can provide insights into the overall adaptive mechanisms of skeletal muscle. This study aimed to investigate the effects of a single session of HIIE with and without hypoxia on transcriptomic response relating to autophagy/mitophagy. Ten healthy males (age: 26 ± 4 years; BMI: 23.3 ± 2) participated in three HIIE sessions randomly order: hypoxia (HY, simulated altitude of 3200m with an oxygen fraction of 0.14), normoxia matched to the relative intensity of hypoxia (NR), and normoxia matched to the absolute intensity of hypoxia (NA). Skeletal muscle samples were collected at fourtime points: before HIIE (B), immediately post-HIIE (P0H), 3 hours post-HIIE (P3H), and 24 hours post-HIIE (P24H). Transcriptomic analysis was employed to identify broad patterns in gene expression related to autophagy/mitophagy in skeletal muscle. The largest number of differentially expressed genes was observed in hypoxia after exercise 24 hours. PCA analysis did not show any significant difference between exercise conditions. Autophagy-animal and mitophagy-animal pathways (not ranked in the top 20) were significantly enriched only in hypoxia 24 hours post-exercise, and the key signaling pathway (mTOR signaling pathway) did not exhibit pronounced changes at all conditions after HIIE. To focus on autophagy/mitophagy-related genes/pathways, I chose 604 genes related to autophagy/mitophagy as target genes to perform differential expression analysis. Again, the largest number of differentially expressed genes related to autophagy/mitophagy was observed in hypoxia after exercise 24 hours. However, recognized autophagy markers such as LC3II, LC3II/LC3I ratio, P62, and mitophagy receptors (NIX, BNIP3) did not show significant differences in expression level in all conditions, and mitophagy receptor FUNDC1 was observed in hypoxia, indicating a lack of robust change in autophagy/mitophagy after a single session of HIIE with and without hypoxia. Similar expression patterns were only observed between different time points under the same exercise condition. Although our result suggests an elevated level of genes related to autophagy/mitophagy pathways under hypoxia exercise, it did not correlate with physiological performance, as a higher percentage of HRpeak and V̇ O2peak, and a higher RPE were observed in both NR and HY compared to NA. Furthermore, the increased gene expression level in hypoxia is likely affected by time points after exercise. Our result suggests that activation of autophagy/mitophagy pathways is unlikely to be responsible for the difference in observed physiological response.

Individual performance can be defined as a players ability to complete the desired task at a substantial quality whereby winning is the ultimate goal. It is very hard to predict performance through training in sports such as Australian Football (AF) given their dynamic and complex nature. Any training monitoring tool or data analysis technique that can provide any insight into match performance or correlation can substantially assist practitioners to get a competitive advantage over their opponents, ultimately leading to short term and long-term success. The primary aims of this thesis were to investigate the extent to which commonly used training monitoring tools demonstrate a relationship with performance when measured from both a subjective and objective perspective. Additionally, it aims to develop a novel approach to visually representing the accumulation of volume across the intensity spectrum for various comparisons throughout the pre-season and in-season phases of the highest level of AF being the Australian Football League (AFL). This new data analysis technique was implemented as it allows for quantification of the distribution of volume accumulated providing more practical insights for practitioners, compared to previous studies whereby the distribution has simply been modelled visually. The specific aims of each study are as follows. Study 1: to determine which characteristic/s if any, have a relationship with match performance identified through Champion Data© rating points and coaches subjective player ratings. Forty AF players from the same club (25 ± 4 yr) were monitored across the 2019 AFL season. Mixed models were used to compare each of these performance ratings with wellness scores and Global Navigation Satellite System (GNSS) metrics such as total distance, surge distance, sprint distance, total high-speed running distance, work rate and maximum velocity. There were two statistically significant relationships identified, wellness energy scores and coaches subjective ratings and wellness soreness scores and Champion Data© rating points. Neither relationship provided meaningful results given both relationships were in a negative direction whereby the findings are not something to be promoted with players. The lack of findings suggest more sophisticated analysis techniques are required investigating suitable metrics that assess performance. Study 2: to implement the new analytical approach and determine the cumulative distribution of volume across the intensity spectrum for an entire home and away season of AF, specifically investigating the accumulation differences between competition levels (AFL vs Victorian Football League (VFL)) and the duration of the match. Thirty-three AF players from the same club (24.4 ± 4.3 yr) were monitored across the 2021 season. The data collected was processed into absolute values with a one minute moving average applied whereby the observations were categorised into 10 m.min-1 and 5 m·min-1·s-1 zones for speed and acceleration respectively and the total distance in each zone was calculated. Quadratic models were then created for each player for each quarter which was then summarised by competition level and quarter of the game for analysis. A clear difference in speed and acceleration between the duration of the match and competition levels were identified. This study identified AFL is played at a higher intensity than VFL and within AF matches, quarter one is the most intense with a decline across the duration of the game in both speed and acceleration. Study 3: to expand on the analysis used in the previous study and apply this technique to examine the distribution of volume across an AF pre-season and investigate the volume during different pre-season phases and training drill modalities. Forty-three AF players from the same club (24.4 ± 4.3 yr) were monitored across the 2022 AFL preseason period, spanning November 2021 to March 2022. The same data analysis technique was used as in study 2, with the only variation being the data analysed. The data was categorised in three pre-season phases, pre-Christmas, post-Christmas and precompetition and four training drill modalities, fundamentals, structure, game plan and conditioning. Training drill modalities were categorised based on predicted work rates identified by the clubs high performance manager. This study identified that the post- Christmas phase exposes players to the greatest volume of pre-season at a high intensity. Within training drills, conditioning drills accumulate the greatest acceleration while game plan drills expose players to the highest intensity. Study 4: to continue expanding on the previously used analysis technique and apply it to in-season, specifically examining the distribution of volume between different match turnaround lengths and whether matches or training accumulate the greatest volume. Forty-three AF players from the same club (24.4 ± 4.3 yr) were monitored across the 2022 AFL home and away season. The same data analysis technique was used as per study 2 and 3, where the comparison was changed to three between match turnaround lengths (8-day) and matches versus training. Within the week, matches regardless of the between match turnaround length accumulate the greatest volume. When examining match volume specifically across the different turnaround lengths, greater impulse is achieved in matches where the between match turnaround length is >8 days. General conclusions: Overall the data from this thesis builds upon the known importance of acceleration in AF and provides quantifiable results that can be utilised in training design, in designing drills based off of these findings. Through understanding the differences between match quarters this may drive a conversation around how training can improve to limit the decline observed between quarter one through four. It can also assist at the sub-elite levels by providing benchmarks for the top speeds and impulse achieved at the elite level, to allow players to train towards. These findings can also confirm training periodisation plans for practitioners whereby, intentions are proven or disproven to be true based on the cumulation observed as the analysis techniques used in these studies can be adapted and applied in many sports.

Regular exercise improves physical fitness and general health via significant muscle molecular adaptations. ATP (energy) and its regulation is critical to metabolic homeostasis. ATP hydrolysis and re-synthesis are typically balanced, disrupted when the rate of ATP re-phosphorylation can no longer support extreme exercise workload, such as high-intensity training (HIT), or production is compromised (metabolic disease). Consequently, purine degradation to terminal purine metabolites (hypoxanthine and xanthine) occurs, which diffuse across the sarcolemmal membrane into the blood converted to uric acid and eliminated by the kidneys. This process generates reactive oxygen species (ROS) that can damage skeletal muscle fibres and the local capillary network. Extreme metabolic insult triggers three possible responses: (1) oxidative stress; (2) endoplasmic reticulum (ER) instability; and (3) hypoxic stress. Cellular oxygen consumption and subsequent ROS production are implicit in the activation of these signaling pathways. ROS stimulates beneficial muscle adaptations associated with regular exercise as per the concepts of hormesis. However, overexposure (acute or chronic) can drive muscle dysfunction and pathology. Elucidation of the signaling pathways altered during intense metabolic stress, e.g. HIT or exercise in metabolically compromised muscle, is needed. Furthermore, supporting cellular metabolism under intense pressure could prevent excessive stress signaling and these negative consequences. Ribose supports purine recovery and biosynthesis, and supplementation may protect metabolically challenged muscles, or prevent formation of the critical signals required to induce beneficial adaptations within skeletal muscle, e.g., ROS. Study 1 examined the molecular stress signaling response to HIT in mice. Following HIT, the levels of Keap1, SOD, and PGC-1α were elevated in the skeletal muscles of female mice, yet the expression of HO-1, Sirt1 and PGC-1, which can be induced by hypoxic stress protein, HIF-1, was upregulated in male skeletal muscle. This indicates female skeletal muscle may be more responsive to oxidative stress, while male muscle may be more responsive to hypoxic stress. Study 2 examined (1) oxidative, metabolic and hypoxic stress signaling in metabolically challenged dystrophic (mdx) mouse model exposed to regular normo-tensive exercise; and (2) whether ribose supplementation could protect muscles from exacerbated myopathy. Ribose treatment enhanced resistance to fatigue and forelimb muscle strength in mdx mice by inducing hypoxia stress signaling. Additionally, we observed that ribose supplementation reduced purine degradation, leading to potential adaptations in ER stress signaling pathways by promoting fat metabolism. Therefore, ribose could be a therapeutic adjunct to treat muscle fatigue in DMD patients and female carriers of the dystrophin gene mutation. Study 3 focussed on exercise performance and stress signaling adaptations with HIT in humans. Impacted heavily by COVID-19, this small pilot study indicated that HIT training induces the Nrf2-mediated antioxidant program and Sirt1which may safeguard muscle cells against oxidative damage caused by exercise. This thesis examined models of stress in healthy and disease states to better metabolic signaling dynamic and influence of ribose supplementation on the adaptation pathways.

Introduction There is a substantial interindividual variability in responses to exercise training, which is influenced by both environmental and genetic factors. However, the influence these genes is relatively unknown. Therefore, we have investigated the association between robust genetic variants, using the Exercise Polygenic Score (EPS), on mitochondrial and physiological response to four weeks of High-Intensity Interval Training (HIIT) in the Gene SMART (Skeletal Muscle Response to Training) study. Thus, the overarching aim of this thesis is to evaluate the association of robust genetic variants using the Exercise Polygenic Score (EPS) in mitochondrial and physiological response to exercise phenotypes to four weeks of High Intensity Interval Training (HIIT). Methods 116 adults from the Gene SMART cohort study completed four weeks of HIIT to measure physiological and mitochondrial responses. Maximal oxygen uptake (VO2max), lactate threshold (LT), and peak power output (Wpeak) were assessed before and after the exercise training. Muscle biopsies were collected before and after 4-weeks of HIIT to assess mitochondrial markers: citrate synthase (CS), cytochrome c oxidase (COX), succinate dehydrogenase (SDH), mitochondrial copy number (mtCN) and mitochondrial health index (MHI). DNA isolated from blood samples was genotyped using the Genome-Wide Genotyping Array, and genotype data was then used to derive exercise polygenic scores for each participant using exercise-related SNPs identified in Chapter 3. Results: We found significant increases in mitochondrial markers CS and COX after the HIIT intervention (p < 0.05), but no changes in mitochondrial health index (MHI). We also found that changes in LT were found to be positively correlated with changes in both CS (r = 0.2, p = 0.014,) and COX (r = 0.19, p = 0.019). Also, significant correlations were found between changes in VO2max and changes in two mitochondrial markers, CS (r = 0.24, p = 0.0025) and SDH (r = 0.20, p = 0.011). Finally, a significant correlation was found between changes in Wpeak and changes in CS (r= 0.24, p = 0.0024). We did not observe significant associations between MHI changes and changes in physiological measurements. We found no associations between EPSs and physiological and mitochondrial markers either before or after four weeks of HIIT. However, we found a significant association between baseline mtCN and the (PPARGC1A) rs8192678 SNP (p= 0.012). We further showed several associations between SNPs and mitochondria factors i) baseline mtCN and rs8192678 (p= 0.021), ii) 4-week change in mtCN and (BIRC) rs6090327 (p

In Australia, 41% of industry injury claims are due to manual handling tasks, costing $14.58 billion annually. In the Australian Army, 78% of physically demanding tasks are considered manual handling, which increases the risk of musculoskeletal injury. Of injuries in active service personnel for the Australian Army, 22% occurred in the trunk [3, 4] with manual handling recognised as the cause for 5% of all injuries [3, 4]. This has led to the need for an exoskeleton system that can support, move and adapt to repetitive, fatiguing tasks. The predecessor to this exoskeleton system is the development of an assist-as-needed control algorithm that will predict when personnel are lifting above their maximum acceptable weight of lift (MAWL), which is indicative of an increased injury risk. This algorithm could also be deployed on a simpler stand-alone wearable device that could assist personnel in reducing risk factors associated with injury due to manual handling tasks, through providing visual or auditory feedback. Laboratory experiments using biomechanical task analysis based on military manual handling protocols were performed with a sample size of 32 participants. Inertial measurement units (IMUs) were used in a six-segment spine model for data collection. The normalised (for time) kinematic output of the IMUs for participants during lift-to-platform tasks were analysed for the relationship between changes in spine kinematics and increasing external load. Statistical parametric mapping was performed to determine significance in the IMU variables. Additionally, polynomial correlation of discrete features were analysed for use as predictive factors of external loading above a participant’s capability which resulted in poor correlation. Machine learning was performed due to its ability to find trends and features in data that may not be apparent via statistical inference. Supervised machine learning algorithms capable of classifying multivariate time series data were compared. The Random Convolutional Kernels (ROCKET) algorithm had the highest accuracy for its ability to classify a high risk (at or above MAWL) or low risk (below MAWL) lift, with a 10-fold cross validation mean accuracy of 91.2 ± 2.7%. A moderate f1-score was maintained through dimensionality reduction of the spine segments and data frames per feature. Reducing the spine segments to one (middle lower thoracic) and data frames to half (50) resulted in a f1-score of 86%. This research contributes an accurate novel predictive model that uses machine learning to classify spine kinematics from IMUs into high and low risk lifts, based on MAWL. In future work, the novel predictive model developed in this thesis will contribute to the development of a stand-alone device providing user-feedback. The model will also be part of an assist-as-needed control system for the development of an active exoskeleton that could provide augmentation to Defence personnel during manual handling. These devices aim to reduce injuries caused by lifting above an individual’s capacity.

The design of practice environments supportive of learning and performance is a contemporary challenge for practitioners in high performance sport. To address such a challenge, the coupling of contemporary pedagogical frameworks, such as the constraints-led approach, with the practical implementation of tools from sports analytics may be beneficial. This thesis explored the measurement and analysis of key task, environmental and individual constraints to guide practice design in professional Australian Football. Across five studies, various analytical techniques were used to evaluate different constraints and their interactions, and determine their effect on athlete and team behaviour. Spatiotemporal player tracking data was first analysed to determine a novel, continuous measure for the constraint of pressure and its influence on performance. Rule association and regression trees were applied to evaluate the influence of environmental, task and individual constraint interactions on athlete skilled behaviour. Univariate and multivariate change point analyses were applied to inform the duration of training activities to support skill learning. Rule association and classification trees were used to evaluate the influence of a numerical constraint manipulation on interacting technical, tactical, and physical team behaviours. Collectively, the findings from these studies not only assist practitioners in the design of practice tasks but show how constraint manipulations may challenge or promote various behaviours in team sports athletes. Moreover, this thesis demonstrates the utility of multivariate analytical techniques in the exploration of constraints interaction in sport. The suitability of such techniques for the measurement of complex and non-linear interactions between athletes, the task and environment, was highlighted. Practitioners can integrate and adapt these analytical tools, in conjunction with the constraints-led approach, to inform the design of practice tasks that facilitate learning and development in high performance sport.

Cricket fast bowling coaches have an abundance of knowledge related to speed production and bowling technique. However, bowling coaches find large gaps of research knowledge related to bowling accuracy. While a consistent pitch length (accuracy) can be achieved from a redundant combination of four ball projection parameters (horizontal velocity, vertical velocity, horizontal position, and vertical position), there is no current research of these relationships. Furthermore, the link between body position and ball projection is unknown. While gait studies have employed kinematic effector units to explain functional biomechanics and control of stance and swing limbs of walking, this type of analysis is rare in sports biomechanics. This motivated this study to divide the whole-body into three effector units (Leg, Trunk and Arm) to analyse how body position effects bowling accuracy. This study aimed to provide preliminary information related to the research gap of the general nature of three-dimensional body position variability and bowling accuracy through investigating pitch length control. A group of 12 sub-elite male fast bowlers volunteered to be biomechanically tested on their performance of 18 good length bowling trials. Ball and body kinematics were captured from a threedimensional motion analysis capture system (250 Hz) setup at an indoor cricket pitch. Each bowler’s set of BR and body effector variables were represented as mean and standard deviation from 18 trials. Ball pitch length was determined via a flight prediction model based on constant acceleration. A multi-linear regression analysis showed that the four release parameters accounted for 79% of pitch length variability, where vertical velocity variability accounted for the greatest variability (90%). A coordinated covariance test did not find any redundant cooperation among the four BR parameters. Therefore, pitch length accuracy appears to be achieved by independent control of vertical velocity. Three-dimensional body effector position vectors were time-normalised to the acceleration phase of bowling (FFC to BR) and presented in each dimension as time-waveform signals. The average variance of the effector profile during the bowling phase was calculated across each bowler’s 18 trials and 11 time-slices. A stepwise linear regression analysis found that variance in Leg-X and Trunk-Z effectors can explain 85% of the variance in vertical velocity variability at BR. A coordination covariance test also revealed that redundant coordination of body effector position exists in the X- and Z-dimension. In summary, the ball vertical velocity at BR is the critical parameter that determines pitch length control, while dimension-relevant variability of the Leg effector and Trunk effector determine vertical velocity variability. In summary, this project identified valuable preliminary information that can readily be translated and applied to coaching methods for bowling accuracy. Furthermore, it offers sound insights for future research to enquire deeper into the nature of fast bowling accuracy.

Ageing is characterised by a simultaneous loss of muscle mass, strength and physical performance (sarcopenia) and bone mass (osteoporosis). These changes in muscle and bone can lead to reduced physical function, increased falls and fractures and a poorer quality of life. Due to the ageing population and increases in sedentary behaviours, sarcopenia and osteoporosis prevalence and associated burdens are predicted to rise. The simultaneous loss of muscle and bone raised the hypothesis that bone and muscle are not only linked anatomically, but also metabolically and chemically. However, it is still not clear how bone-derived factors are involved in this crosstalk. Bone is an endocrine organ, releasing hormones affecting distant tissues and organs. Osteocalcin (OC) is the most abundant non-collagenous protein in bone. Its total serum levels (tOC) are used clinically as a bone turnover marker (BTM). The undercarboxylated form of OC (ucOC) is considered bio-active, involved in energy metabolism and possibly muscle mass maintenance and strength, at least in rodents. Evidence from human studies is limited and contradictory, in part because most research has focused on tOC, rather than the ucOC. Exercise improves muscle and bone mass, as well as muscle strength, while inactivity has deleterious effects on both organs. Consequently, exercise is a cornerstone approach to maintain and preserve musculoskeletal health in adults, and can be used as a tool to investigate bone-muscle crosstalk. The primary aim of this PhD thesis was to identify the normal range of ucOC across the adult human lifespan, and to explore whether ucOC and other BTMs are related to muscle mass, strength and physical performance. I also explored whether acute exercise can affect the relationships between ucOC and muscle function (strength and physical performance) in older adults. The specific aims were: Study 1: to develop age-based reference ranges for OC and its forms and ratios in healthy adult men. Overall, 236 adult men participated in the study (18 to 92 years old). Serum samples were analysed for tOC and ucOC (using the hydroxyapatite binding method) and carboxylated OC (cOC), ucOC/tOC and cOC/tOC ratios were calculated. Ageing was associated with a “U” shaped pattern for tOC, cOC and ucOC levels. The ucOC/tOC ratio was higher, while cOC/tOC ratio was lower, in men of advanced age, demonstrating that OC ratios may be better measures than the absolute values to identify age-related changes in OC. Study 2: to test the hypothesis that the serum ucOC absolute value and ucOC/tOC ratio are associated with muscle function and long-term risk for falls-related hospitalisations using a large longitudinal dataset (15 years) in older women (n=1261, mean age 75.2±2.7 years). In older women, a higher ucOC/tOC ratio was related to poorer physical function, including the long-term decline in physical function and increased risk of falls-related hospitalisations. Early The identification of women at higher risk for functional decline using the ucOC/tOC ratio may enable prevention and intervention strategies to occur early, reducing future risk for injurious falls. Study 3: to perform a systematic review to examine the effects of acute exercise on BTMs in adults over the age of 50 years and identify whether BTM responses are determined by exercise mode, intensity, age and sex. Thirteen studies were included: eight in middle-aged adults (n= 275, 212 women/63 men, mean age= 57.9±1.5 years) and five in older adults (n= 93, 50 women/43 men, mean age= 68.2±2.2 years). Eleven studies included aerobic exercise (AE) (7 middle-aged/4 older adults) and two included resistance exercise (RE) (both in middle-aged adults). AE increased C-terminal telopeptide of type I collagen (CTX), alkaline phosphatase (ALP) and bone-ALP in middle-aged and older adults. AE also increased tOC in middle-aged men, and procollagen I carboxyterminal propeptide (PICP) and cross‐linked carboxyterminal telopeptide of type I collagen (ICTP) in older women. In middle-aged adults, RE combined with impact exercise (jumping) had no effect on tOC or BALP, but led to a decrease in CTX. Jumping alone increased P1NP and tOC in middle-aged women. Acute exercise is an effective tool to modify BTMs, but the response appears to be specific to exercise modality, intensity, age and sex. Study 4: to test the hypotheses that a) at baseline, serum ucOC and other BTMs are associated with muscle function, b) acute exercise can alter ucOC and BTMs and c) muscle function at baseline is related to the acute exercise responses of these biomarkers. A total of 35 older adults (25 females/10 males, 72±6 years) participated. The baseline assessments included: body composition, handgrip strength and a physical performance test (PPT) (gait speed, TUG, time to climb and descend 10 stairs). Leg muscle quality (LMQ) and stair climb power (SCP) were calculated. Participants performed in a randomised order a single session of 30 mins AE (cycling at 70% of peak heart rate) and RE (leg press at 70% of one repetition maximum and jumping regimen). At baseline, higher muscle strength was associated with higher P1NP and better physical performance (lower PPT score). Similarly, higher SCP was associated with higher P1NP and the betafragment of CTX (ß-CTX) (p

The collective positioning and behaviours of elite netball players are an important concept to understand, as they relate to the tactical plans and overall performance of a team. However, this is currently an under-researched field in most indoor team sports owning to previous limitations in indoor tracking systems. Thus, this thesis investigated the use of collective tactical variables in elite netball matches and training, captured using a local positioning system (LPS). Collective tactical variables were used to explore collective movement and positioning behaviours in matches, examining the more successful movement patterns and how these varied in small sided games (SSGs), commonly used in training. Study 1 assessed the validity of the Catapult T6 LPS to measure inter-unit distance. Eight recreationally-active, female indoor team-sport players completed a circuit comprised of seven movements (walk, jog, jump, sprint, 45º change of direction and shuffle), on an indoor court. Participants wore a receiver tag (ClearSky T6, Catapult Sports) and seven reflective markers, allowing for comparison with the reference system (©Vicon Motion Systems, Oxford Metrics, UK). The criterion validity against the reference system resulted in a root mean square error of 0.20 ± 0.05 m. Additionally, analysis was conducted on five distance bands to assess the association between distance and inter-unit distance validity. The resulting mean bias presented an overestimation for the LPS in all distance bands, while the percentage of variance accounted for remained stable for all distance bands, excluding the > 20 m band. These results suggest the Catapult T6 LPS has acceptable accuracy - within 20 cm - for measuring inter-unit distance, as such opening up opportunities for further research into collective tactical variables for indoor team sports. Study 2 of this thesis used the validated ClearSky LPS to explore collective tactical variables during seven elite netball matches at the team and sub-group level. Collective tactical variables were explored during attacking and defending possessions and included; inter-player distance, width, length, surface area, centroids, width-per-length ratio and stretch indices. During attack, the team and all sub-groups adopted greater lateral dispersion between players, while on defence there was generally greater longitudinal dispersion. Additionally, greater irregularity (measure of predictability within time-series data) was observed in sub-groups most present and active in the play (forwards on attack, defenders on defence). Study 3 followed on from the results of the previous study, by modelling the effects of collective tactical variables and contextual predictors on scoring probability in elite netball. The effects were estimated via a logistic-regression version of the generalized linear mixed model, with adjustments for several predictor variables. Extremely short possessions were associated with decreased scoring probability, while lateral dispersion negatively affected the probability of scoring for both teams. These results indicated that, in general, it is more favourable to maintain positioning and ball movement in the middle of the court. Finally, Study 4 focused on SSGs, which are commonly used in training to replicate match outputs. The collective tactical variables of four attacking and four defending elite netball players were analysed to compare three SSG conditions: 4v4, 5v5 and 6v6. Two, 20-min SSGs for each condition were analysed using a linear mixed model. The irregularity of width-per-length ratio of the four attacking and four defending players was greatest in 4v4, the attackers’ inter-player distance and length variability was greatest in 5v5, while the four attacking players had the smallest mean surface area in 6v6. This thesis provides an exploration into the use of collective tactical variables in elite netball. As a combination of studies, this thesis is an important investigation into methods available for performance analysis research in netball and other indoor team sports. The key findings showing tactical analysis via collective tactical behaviours hold useful information for elite netball. Of interest, different patterns of positioning and movements exist between attacking and defending phases of play and sub-groups. All sub-groups displayed increases to their lateral dispersion during attack and active sub-groups presented greater irregularity, aligning with the positional dependencies and restrictions of netball. Also, certain collective tactical variable outputs are more successful than others, most prominent being shorter possession associated with decreased score probability, suggesting maintaining ball possession early is a key driver in score probability. Finally, in training SSGs differ tactically with different number of players included in the drills, with lower numbers causing defenders to exhibit greater irregularity while for greater numbers, players were closer to their teammates, both results displaying specific coaching principles that can be trained.

Previous research has found mental imagery (MI) to be an effective tool for enhancing performance and skill learning across a range of contexts from sport to music and dance. The beneficial effects of MI are often contextualised through the functional equivalence hypothesis, which proposes that MI and physical practice (PP) activate similar neural mechanisms. Therefore, MI research has often emphasised replicating critical aspects of PP in MI to maximise functional equivalence. Despite this, the MI research literature has rarely investigated the application of key skill acquisition principles commonly utilised in PP to MI. One approach to skill acquisition strongly influencing our current understanding of how to develop skill is nonlinear pedagogy (NLP). NLP draws on key principles of ecological dynamics, proposing that skill development is nonlinear, and emerges as the result of dynamic learner-environment interactions. Captured this way, implementation of NLP design principles such as constraint manipulation, representative design, information- movement coupling, and movement variability are proposed to facilitate the development of adaptable, individualised movement solutions. Given the emphasis on creating MI practice conditions that mimic PP, using a NLP informed approach in MI may help facilitate the effective development of adaptable, individualised skills that can deal with changing performance contexts. The inclusion of non-sport related skills (e.g., finger tapping) and other psychological techniques in previous MI reviews made it difficult to ascertain how MI directly attributes to skill development in sport. Therefore, Study 1 attempted to address this gap in the literature through a meta-analytic review of MI, focusing on sport-specific-motor skills. From the 36 studies reviewed, it was found that MI has a significant positive effect on the development of sport-specific-motor skills (g = 0.476). Further analysis revealed MI combined with PP to be most effective (g = 0.579). Skill complexity, MI delivery type (i.e., MI combined or independent of PP), and performance measures were found to moderate the efficacy of MI interventions. A key finding that has helped to further understanding about the impact of MI interventions was the identification of skill complexity as moderating MI effectiveness. Results indicated that MI interventions practicing simple skills are significantly more effective relative to more demanding, complex movements. However, MI research investigating complex sport-specific-motor skills was lacking (5/58 analysed effect sizes). The need for further MI research on complex sport-specific-motor skills formed the rationale for selecting the power clean (PC) – a complex Olympic weightlifting skill – in Study 4 of this thesis. An ecological dynamics perspective of skill acquisition in PP was presented in Study 2, highlighting the importance of adaptability in skill development and the relevance of such a perspective to MI was discussed. The applicability of NLP design principles to MI interventions was presented alongside practical examples of how these principles could be integrated alongside existing MI guidelines. Key considerations included the incorporation of movement variability through task constraint manipulation, inclusion of critical aspects of performance environment (e.g., defensive pressure and other team members), and the use of movement outcome focused instructions. This study provided the theoretical foundation to further examine the application of NLP to MI interventions. Based on the review of NLP research in Chapter 2, the influence of a NLP approach for skills that emphasise movement form for performance (e.g., Olympic weightlifting) was identified as an area for further investigation. Specifically, preliminary case study evidence was identified using a constraints-based approach (key methodological influence on NLP) to the PC (e.g., Verhoeff et al., 2018) which presented promising results for the use of ecological dynamics informed approaches (i.e., NLP). Therefore, the effectiveness of NLP relative to linear pedagogy (LP) practice was investigated in Study 3. This study involved beginner learners to engage in either NLP or LP practice of the PC over a 4-week intervention. Contrary to predictions, exploratory behaviour (i.e., presence of movement variability) was not significantly different between NLP and LP conditions. However, equivalent improvements between conditions in performance accuracy (i.e., FxD; forward barbell displacement) were observed. This suggested that pedagogical approach may not be a precondition for adopting a particular technique, and inherent individual and task constraints may require learners to self-organise behaviour to develop an individual task solution. Importantly, deviation from an instructed technical model does not appear to lead to less efficient performance. Such a finding has important implications for a practitioner’s overall learning philosophy. Even in activities considered to rely upon a specific movement form (i.e., PC), practitioners might want to distinguish between techniques that are effective and movement patterns that look correct. That is, the effectiveness of the movement for producing specific outcomes may be more a more important consideration than reproducing the ideal ‘aesthetic’ or movement style (i.e., what the movement looks like). Study 4 aimed to investigate the influence of a NLP approach to MI in relation to a LP practice approach to MI with beginners learning the PC. The same design was implemented as Study 3 (i.e., practice sessions, instructions, reps, and 3-D motion capture and horizontal barbell displacement). Like Study 3, no significant differences were observed in exploratory behaviour and equivalent improvements in performance accuracy (RxD; rearward barbell displacement) for both conditions. Consistent with the Study 3, these findings suggest that establishing a movement pattern that achieves overall performance goals (e.g., reduced forward or backward barbell displacement) is more important than replicating a prescribed technique that looks correct. Considered with the results of Study 3 these results suggest an equivalence in training-related improvements, highlighting that it may be possible to reproduce similar behavioural adaptations observed (to a lesser magnitude) when using NLP or LP approaches in MI. In summary, the aim of this thesis was to enhance our understanding of applying a NLP approach to MI intervention design. Therefore, this thesis did not aim to investigate whether NLP was ‘better’, but rather to provide preliminary findings and hopefully stimulate further discussion about incorporating established skill acquisition principles from PP into MI. It is not definitive whether NLP provides further benefits to skill development over and above LP practice, however, the overall improvement in performance outcomes in Studies 3 and 4 suggest it is a legitimate consideration for future interventions. The lack of significant differences between LP and NLP conditions suggests that despite being prescribed a specific movement form, learners may search for more individually appropriate techniques, and importantly, these deviations are not necessarily detrimental to overall performance.

High-performance athletes are known to be at risk of adverse physical and mental health outcomes related to the extreme performance demands they face. Australian high-performance sport has traditionally been reactive to the healthcare needs of athletes who present having experienced compromised health, whether that be in the form of an illness, an injury, or a mental health challenge. Recently, there has been a move toward implementing limited health promotion approaches addressing identified health risks in the current athlete/sport environment. At present, there is no theoretical framework to help those working with this highly selective group to promote the development and maintenance of optimal holistic health. Moreover, the early and ongoing development of optimal holistic athlete health has not commonly been viewed as a performance requirement within sporting organisations. This research aimed to understand health and its determinants for Australian high-performance athletes from the perspective of a representative sample of expert health practitioners who currently work or have recently worked in high-performance sports healthcare. Determinants of health are factors that influence how likely we are to stay healthy or become ill/ injured. Understanding these determinants will help researchers and practitioners develop a theoretical health promotion framework that may be applied broadly across Australian high-performance sport. This study used a mixed-methods approach, including a Delphi survey and subsequent semi-structured one-to-one qualitative interviews to derive a consensus on athlete health determinants and the factors which influence these. The purpose of the qualitative interviews was to explore, more deeply, the experiences, beliefs and thinking behind participant responses to the questions posed in the Delphi survey, to add both nuance and context to those responses. The Delphi survey was conducted at two time points over a period of three months. Descriptive statistical analysis of the Delphi survey results demonstrated that there was general recognition of the relevance of the World Health Organisation definition of health and its determinants to the health of high-performance athletes, a finding that is not usually evident in the literature, nor reflected in the provision of health services to this population cohort. Thematic analysis of the Delphi survey data, using a socio-ecological theoretical lens, revealed a need to consider a broader range of influences on athlete health than those that relate specifically to participation in high-performance sport. These include geographic isolation, access to appropriate health services, ethnicity, and socio- economic status. The interview findings revealed that there is a need to apply additional consideration to factors that can be categorized as social determinants of health in healthcare planning for, and management of, Australian high-performance athletes. This study's findings will help form a theoretical framework for Australian high-performance athlete health. Specifically, this framework would address the need for sporting organisations to create and provide health-promoting environments for their athletes and to support athletes in developing their capacity to manage better the impact of the health stressors to which they are exposed, as identified by the high-performance healthcare experts surveyed in this study.

The objectives of this thesis were to establish the validity of commercially available rowing instrumentation systems for the measurement of oar angles and power to inform the interpretation of these measures, extend knowledge regarding the prediction of on-water rowing race performance and the contribution of measures of rowing technique and boat acceleration to rowing performance. From a practical perspective, this thesis aims to enable coaches and sport scientists to identify specific areas where performance improvements can be attained in the athletes they work with. Studies One and Two established the concurrent validity of rowing instrumentation systems for measures of oar and power, respectively. Concurrent validity was acceptable for Peach systems for oar angle (trivial to small but unclear systematic bias, and trivial to small random errors) and power per stroke (possibly or likely trivial random errors of -3.0 to -16%). EmPower systems did not have acceptable validity for oar angle (trivial to small but unclear systematic bias, and moderate to extremely large random errors) or power per stroke (likely or decisively substantial random errors of 9.7 to 57%). Only power was assessed in Weba and Concept2 systems as catch and finish oar angles are not measured by these devices. The random error associated with power per stroke was 61 to 139% for Weba and -28 to 177%) for the Concept2. Systematic bias in mean power was negative for all devices (Concept2, -11 to -15%; Peach, -7.9 to -17%; EmPower, -32 to -48%; Weba, -7.9 to -16%). Study Three investigated the contributions of power, stroke rate, headwind, technical efficiency, race conditions, and stroke-velocity variability in 45 rowing race performances. The unexplained prediction error was 0.35 to 0.55% across the four boat classes assessed. Effects on race velocity were extremely large for mean race power, small to large for mean stroke rate in singles, large to extremely large for headwind, trivial to extremely large but unclear for technical efficiency, very large or extremely in singles for race conditions, and small to trivial but mostly unclear for stroke-velocity variability. Specific measures of rowing technique from the Peach system were investigated in Study Four by evaluating their individual relationships with rowing velocity. Substantial relationships with velocity were found between most variables before adjustment for power and stroke rate, but effect magnitudes were reduced after adjustment for power and stroke rate. The greatest modifying effects were found for stroke rate, mean and peak force, and power output before adjustment, and for catch angle after adjustment for stroke rate and power. Study Five explored relationships between boat acceleration profile and rowing performance. Several measures of acceleration magnitude and jerk had substantial effects before adjustment but were reduced in size after adjustment for stroke rate and power. Substantial effects were found for maximum negative drive and peak drive acceleration magnitudes, and jerk in the early-to-mid drive and late recovery phases after adjustment for stroke rate and power. The findings of this thesis show Peach instrumentation systems have adequate reliability for stroke-to-stroke assessment of oar angles and power output. Race performance can be predicted from power, stroke rate and technical efficiency. Stroke rate, power, force, and catch angle are key areas where improvements in performance can be attained, as are changes in boat acceleration profile reflecting the late recovery, catch placement and early drive phases of the stroke.

Natural performance enhancement, muscle hypertrophy and muscle rehabilitation on an expedited timeframe is highly coveted in sports and fitness industries. Heat stress (HS) can enhance muscle hypertrophy and strength when used in combination with resistance exercise (RE), however, the additive effects of HS on power, speed and agility have not been investigated. Furthermore, the efficacy of previously tested heating methods in achieving muscle heating is highly variable and inconsistent, which probably contributes to the inconsistent effects of heating reported in the literature. Therefore, primarily, this thesis aimed to investigate the effects of HS on performance adaptations to RE. Secondarily, how the cellular signal transduction pathways regulating skeletal muscle adaptations to RE are altered with long term concurrent HS were also investigated. Thirdly, we aimed to develop a reliable localised heating method. We hypothesised that full body HS applied concurrently to heavy progressive RE may improve upon the phenotypic and molecular muscle adaptations to RE. In the first study, we aimed to develop a localised heating method that is capable of raising core muscle temperature above 38-5-39°C, by testing three different models of localised heating. However, none of the models were able to raise core muscle temperature (CMT) to the desired levels. Therefore, in the second study, eighteen recreationally active males were assigned to two groups, HEAT (n=8, 40°C, 30% RH), and CON (n=10, 23°C, 20% RH). Each group undertook an identical, ten week, full body RE program three days a week. Peak core body temperature (HEAT 38.18 ± 0.27 °C; CON 37.97 ± 0.32 °C), as well as vastus lateralis muscle temperature 3.5 cm under the skin (HEAT 36.79 ± 1.55 °C; CON 35.94 ± 1.51 °C) were measured. Strength, peak force, speed, agility and body composition (DXA) were measured pre-, mid (week 5), and post-intervention. Muscle biopsies were obtained from the vastus lateralis pre-intervention at rest, one hour and 48 hours post the first resistance training session. An identical biopsy trial was performed ~72-96h after the last training session of the intervention. In study three, the muscle samples at each time point were analysed via western blots for key markers of muscle protein synthesis. Fibre cross sectional area (CSA), satellite cell (SC) and myonuclear density were quantified via immunofluorescence. In study four, muscle samples were analysed for the heat shock protein (HSP) response via western blots. Study five investigated the mitochondrial and angiogenic adaptations via western blots and capillarisation response via immunofluorescence. Ten weeks of RE improved lower body strength and relative upper body strength, however had no effect on power, agility or speed. Lean muscle mass, fibre CSA, SC content and myonuclear density improved in response to RE. Concurrent full body HS applied at 40°C failed to increase CMT in the vastus lateralis. Therefore, full body HS applied at 40°C did not improve upon performance, molecular or phenotypic adaptations to RE.

This thesis examined the within-person and between-person associations between physical activity, affect and satisfaction with life in adolescents’ daily lives. Understanding these associations is important because integral affect experienced during physical activity may predict future participation in physical activity behaviours. Additionally, understanding if adolescents who are more active experience more positive incidental affect and satisfaction with life is important to illuminate the benefits of physical activity on wellbeing. Despite growing recognition of the within-person and between-person associations between physical activity, affect, and satisfaction with life, little is known about mechanisms that may moderate or mediate these relationships. Therefore, the overarching aim of this thesis is to investigate the mechanisms that may influence these associations. A mixed-methods design was used to achieve this aim. Overall, 125 adolescents participated in an ecological momentary assessment (EMA) study. They were asked to report their momentary behaviours, physical environment, social context, and affect on multiple occasions each day for four-days. Additionally, participants were asked to complete a daily diary each evening to report on their daily satisfaction with life and domain specific physical activity. Participants also wore an accelerometer to assess device measured levels of physical activity over the four days. Additionally, 15 participants completed semi-structured interviews to identify factors that lead to experiencing more favourable affect whilst active. Results from the EMA study showed that participants who engaged in more recreational moderate-to-vigorous physical activity (MVPA) on average than others reported more positive valence, greater energetic arousal, and less tense arousal. Additionally, adolescents reported feeling more energetic, but also more tense when participating in more recreational MVPA than usual. No domains of MVPA had a within-person association with valence. Considering the social context and physical environment, adolescents reported more positive valence and energetic arousal whilst outdoors than indoors and more energetic arousal when with others than alone, indicating there may be some additional benefits to being physically active in outdoors and with other people. Additionally, results showed that being active was only inversely related to tense arousal when completed outdoors. Results from semi-structured interviews added to these findings and indicated several social (e.g., doing physical activities alone or with others), environmental (e.g., doing physical activities outdoors), and cognitive factors (e.g., finding activities fun and enjoyable) were perceived by participants to influence their affective experience whilst active. Finally results from daily diaries showed that adolescents were more satisfied with their life on days when they accumulated more device-measured overall, light, and moderate-intensity physical activity than usual. There was also an indirect between-person association between leisure-time physical activity and device-measured vigorous intensity physical activity through energetic arousal. The results from this thesis show that recreational physical activities have the strongest between-person association with subjective wellbeing in adolescents. Physical activity guidelines for mental health that focus on recreational physical activities may be warranted. More research is still needed, however, to fully understand the mechanisms that explain why recreational physical activities are more strongly related to wellbeing. Results also show that the context in which physical activities occur can influence the within-person association with affect. Therefore, it is important to consider the physical environments and social contexts physical activities are participated in when examining the within-person association between physical activity and affect.

Physical activity is the most effective intervention to enhance health and prevent chronic diseases, such as obesity, type 2 diabetes, and cardiovascular disease. Studies and consortiums have aimed to understand the underlying molecular mechanisms that bring about a healthier phenotype with exercise training, and growing evidence suggests that epigenetic changes, which are molecular modifications to the DNA, play a large role in regulating exercise training adaptations. DNA methylation is the most widely studied epigenetic modification in exercise training studies, as it has been shown that both acute and maintained exercise (i.e. training) induce changes in the DNA methylome and subsequent gene function in human skeletal muscle. However, to date, studies identifying skeletal muscle epigenetic adaptations to exercise training have not investigated whether there is a sex-specific effect, despite skeletal muscle being one of the tissues with the most sex-biased gene expression. A majority of the animal and human studies that have guided our understanding of the underlying molecular adaptations to exercise training have either included only males or pooled males and females together without considering potential sex differences. However, biological sex has been identified as a confounding variable across many biological disciplines, and sex- specific analysis can be critical to the interpretation, validation, reproducibility and generalizability of research findings [1]. Thus, the overarching aim of this thesis was to investigate the sex-specific epigenome-wide response to exercise training. Sixty-five healthy males and females (females n = 20; males n = 45) from the Gene SMART (Skeletal Muscle Adaptive Response to Training) study completed four weeks of high-intensity interval training (HIIT) to assess sex-specific training-induced DNA methylation changes. This thesis involved adding the female cohort to the already existing male cohort, of which most of the participant data was collected prior to the commencement of this thesis. Participants underwent a four-week control period prior to commencing the training intervention. To determine whether training induced similar changes in physiological fitness in males and females, three measurements were assessed – maximum oxygen consumption (VO2max), peak power output (PP), and lactate threshold (LT) – at three time points (before control, before and after the HIIT intervention). To assess sex differences in DNA methylation and other molecular measurements (fibre type proportions and gene expression), skeletal muscle biopsies were collected at each time point and analysed with the Illumina HumanMethylation EPIC array. In Chapter 3, we have shown that there are 56,813 differentially methylated positions (DMPs) in the autosomes of male and female skeletal muscle at rest (false discovery rate [FDR] < 0.005), using a large scale meta-analysis of three independent cohorts (Gene SMART, FUSION, and GSE38291) comprising 369 individuals. These DMPs were mostly hypomethylated in males (94%), and were annotated to 10,240 differentially methylated regions (DMRs) and 8,420 differentially methylated genes (DMGs). Gene set enrichment analysis (GSEA) revealed enrichment of sex-differential methylation among muscle contraction, anatomical structure, and metabolism related pathways. Overlapping DMGs with genes known to have sex-biased skeletal muscle expression (differentially expressed genes [DEGs] from GTex), revealed a significant enrichment of DEGs among DMGs. We confirmed over-representation of DEGs among DMGs with transcriptomic data in an additional cohort (FUSION) which was also included in the DNA methylation meta-analysis. Lastly, using qPCR, we verified gene expression sex differences of three top genes identified from the differential methylation and expression analysis in an additional cohort included in the DNA methylation meta-analysis (Gene SMART). In Chapter 4, we investigated the underlying biological factors contributing to the observed sex differences in basal skeletal muscle DNA methylation. Using a meta-analysis approach in the Gene SMART and FUSION cohorts, we have shown that type I muscle fibre proportions were associated with DNA methylation at 16% of sex-biased DNA methylation loci. We found that circulating sex hormone levels (estrogen, testosterone, free testosterone, and sex hormone-binding globulin) in the Gene SMART cohort were not associated with differential methylation at the sex-biased DNA methylation loci. Lastly, we identified that the meta-analysis sex-DMPs were enriched for transcription factor binding sites (TFBSs) of 41 transcription factors (TF) , as previously established by uniform processing of multiple ChIP- seq data sets, including sex hormone-related androgen (AR), estrogen (ESR1), and glucocorticoid (NR3C1) receptors. In Chapter 5, after elucidating the basal skeletal muscle DNA methylome sex differences and their biological contributors, we investigated whether there are sex differences in exercise training-induced DNA methylation changes. First, we found that both males and females improved the physiological fitness measurements PP and LT, but not VO2max, in response to the HIIT, with no sex differences in the degree of the responses. We identified 1,261 CpGs whose methylation changed after four weeks of HIIT at a stringent FDR threshold < 0.005. We found no sex-specific DNA methylation changes after four weeks of HIIT (sex- by-training interaction) at a stringent FDR threshold < 0.005. A global examination of all the statistical tests performed genome-wide did not reveal an inflation of near zero p-values, suggesting that males and females do not differ in their epigenetic response to four weeks of HIIT. Given the relatively short training intervention, we then aimed to investigate whether there were sex differences in DNA methylation associated with cardiorespiratory fitness (CRF), an indicator of lifelong physical activity levels. We found 27,987 DMPs associated with CRF (FDR < 0.005), and no sex differences in the association between CRF and DNA methylation. The experimental design and meta-analysis of this thesis provided large-scale epigenome-wide insight on skeletal muscle epigenetic sex differences, and elucidated the role of DNA methylation in exercise training adaptations in both males and females (Chapter 5). It yielded a comprehensive understanding of the profound sex-specific skeletal muscle DNA methylation and transcriptomic profiles (Chapter 3) and the underlying biological factors (Chapter 4) that distinguish male and female skeletal muscle DNA methylomes. Specifically, muscle fibre type proportions were associated with sites displaying sex differences in DNA methylation; nonetheless, the vast majority of loci that exhibit sex-biased DNA methylation differ regardless of sex differences in fibre type proportions. In addition, although circulating hormones were not associated with sex-differential DNA methylation, the enrichment of hormone-responsive TFBSs suggests that hormones underlie a portion of the DNA methylation sex differences in skeletal muscle. However, the influence of other biological factors, such as the sex chromosomes, on the sex differences observed in the autosomal DNA methylome remains to be determined. Lastly, despite the plethora of sex differences in the skeletal muscle DNA methylome at rest, the DNA methylomes of males and females responded similarly to exercise training as well as lifelong physical activity. These novel findings shed light on the epigenetic response of skeletal muscle to exercise training in healthy males and females. Integrating the DNA methylome with downstream -omics, such as transcriptomics, proteomics, and metabolomics, will further elucidate the pathways and networks involved in the skeletal muscle response to exercise training as well as any sex-specific adaptations. Future studies should include males and females in exercise training studies, take sex and other sex-related factors into consideration in study design and analysis, as well as integrate other OMIC layers to better characterise the skeletal muscle response to exercise training in humans.

Australian Football requires physical and skilled output from its participants for more than ninety minutes of play. In both research and practice, physical output is typically described using aggregate parameters extracted from wearable technologies. Parameters include volume measures (eg., total distance), work rates (volume expressed relative to time, eg. metreage per minute) and output bands, which bin either accelerations or velocity into a smaller number of thresholds. Similarly, skilled output may be described using coaches’ ratings, player rankings and counts of skilled actions, termed involvements. Involvements refer to skilled actions when players are both in possession and not in possession of the ball. These parameters are typically aggregated across pre-set windows, including stints, quarters, and training drills. However, there are periods of altered physical and skilled output within training drills and stints, which are not captured by aggregate parameters. It is also difficult to determine when output meaningfully changes within sessions using these aggregate parameters. Consequently, it is difficult to use aggregate parameters to inform time-based decisions, including substitutions and stint-to-rest, and training drill length prescription. The aim of this thesis therefore was to develop an alternative method to aggregate parameter profiling, which can identify changes— either increases or decreases-- in physical and skilled output within training drills and matches. Study One quantified the relationship between physical output, skilled output and stint duration in elite Australian football matches. Physical output was quantified using aggregate parameters, extracted from Global Navigation Satellite Local Positioning System devices. Skilled output was quantified using individual player involvements. Random effect models showed negative relationships between duration, high intensity running, and involvements per minute. Metreage per minute had a positive relationship with involvements per minute for most players. Three conditional inference trees were computed. These models described the impact of factors, including round (ie., game number within a season) and rotation number, and how individuals react to outputs, along with a general set of thresholds for the data. All models demonstrated a weak relationship between physical, skilled output and time. This suggests that wearable technology data and notational analysis feeds could be analysed differently to improve their use in team sports. Study Two proposed a combined time-series/frequency domain approach to profiling physical and skilled output in team-sport. A binary segmentation change point algorithm was applied to the velocity time-series, collected via wearable technologies of Australian football players during matches. This method overcame the need for pre-set aggregation windows by identifying different segments of physical output through the mean and variability of velocity. Spectral and involvement features were extracted for each segment to describe physical and skilled output respectively. Spectral features were able to describe aspects of output that are not captured using aggregate parameters. For example, spectral kurtosis may describe whether physical output is continuous or intermittent. Between five and seven change points were able to give more insight into physical and skilled output than aggregate parameters, whilst identifying sufficiently different segments of play. Study Three applied the time-frequency approach of Study Two to match profiling in team- sport. This study demonstrated how a time-frequency approach may identify differences in physical output between matches, that are not apparent from aggregate parameters. Additionally, the time-frequency approach was able to identify changes in physical and skilled output within matches. Alongside the change-point algorithm, k-means clustering allowed for segments of movement to be classified through both their time elapsed within a match, and their physical and skilled output. These methods could therefore be used, to increase the specificity of load monitoring and physical activity prescription in team-sports. Study Four illustrated how a time-series/frequency-domain can be applied to physical output to assess the sequence, specificity and difficulty of team-sport training drills. By condensing velocity data from training drills into a similarity metric relative to match segments, a drill sequence resembling physical output at differing points of a match was generated. This study identified challenge points for each drill, where the mean and variance of velocity within training drills changes. The location and features of challenge points varied substantially by drill. Aggregate work rate parameters may therefore misrepresent the influence of training drill length on physical output. Movement paths were further analysed to explore how players accrue total volume measures such as total distance. These movement paths may reveal differences in physical output between training drills to match outputs, despite similar aggregate parameters. This thesis demonstrated how a time-frequency analysis of physical and skilled output may increase the sophistication of match and training drill profiling in team-sport. The methods presented in this thesis can identify periods of high physical output late in a match and the movement paths completed by athletes, with differences in physical output between matches. This information may assist practitioners to identify difficult matches (ie., matches with high physical outputs), without relying on typical aggregate parameters. These methods may also increase the specificity of training drill prescription to match outputs. The methods presented may also inform training considerations that are not addressed with aggregate parameters, including training drill sequence and duration.

Sports compression garments (SCG) are commonly used in athletic applications to improve recovery from exercise. Although the underlying mechanisms are not yet fully understood, they may be closely associated with alterations in blood flow, consistent with that reported in therapeutic medicine. As such, SCG have been implicated in increasing venous and muscle blood flow, and subsequently reducing symptoms of exercise-induced muscle damage (EIMD). However, research investigating the effects of SCG on blood flow, particularly during the post-exercise period, is limited. Chapter 2 systematically reviewed and analysed the effects of SCG on peripheral measures of blood flow (i.e., venous and muscle blood flow) at rest, during, immediately post, and in recovery from a physiological challenge. From the 19 studies included in this meta-analysis, SCG appear to enhance venous and arterial measures of peripheral blood flow during and in the recovery of a physiological challenge. Also, this chapter highlighted that further research should aim to address the limitations of current compression research by reporting the pressure of the SCG, the blinding of participants, and assessing changes in blood flow during recovery. The first experimental study of this thesis (Chapter 3) aimed to comprehensively investigate the effects of three different SCG types (socks, shorts, and tights) on resting markers of venous return, muscle blood flow and muscle oxygenation. Although sports compression tights were the most effective garment, all SCG types positively affected lower- limb blood flow. Thus, SCG may be a practical strategy for augmenting blood flow in the lower limbs at rest. The next study of this thesis (Chapter 4) aimed to investigate the effects of SCG on blood flow post-eccentric resistance exercise, and the influence on aspects of muscle recovery. This study also aimed to determine if the placebo effect is responsible for the improved exercise recovery associated with SCG use post-exercise. This was achieved by incorporating a placebo intervention that participants were informed was as effective as SCG for recovery and matching belief between the SCG and placebo conditions. Compression tights used post-exercise appear to increase blood flow and enhance psychological and performance indices of exercise recovery compared to both placebo and control conditions. These findings highlight that the benefits of SCG are likely not due to a placebo effect. The final study of this thesis (Chapter 5) investigated the effects of SCG on skeletal muscle microvascular blood flow by using contrast-enhanced ultrasound (CEU), a novel technique in compression research. In addition, macrovascular blood flow (i.e., femoral artery), muscle oxygenation, and exercise performance were measured before, during, and following repeated-sprint exercise (RSE). Compression tights attenuated muscle microvascular blood flow following exercise, but a divergent increase in femoral artery blood flow was also observed. However, despite these compression-induced alterations in macro and microvascular blood flow, there was no difference in exercise performance with SCG. Based on this thesis's findings, SCG appear to benefit macrovascular blood flow, with a divergent effect on microvascular blood flow. Also, compression-induced increases in blood flow for up to 4 h post-resistance exercise coincided with improved muscle recovery.

Worldwide the population is ageing, and with this there is increased cost of age- related conditions to both the medical system and, importantly, the individual. Falls are a concern amongst this demographic due to potential loss of independence, and even death. Knee osteoarthritis (OA) is a common chronic condition in older adults which increases the risk of falling. Despite 40-50% increased falls risk in people with knee OA, little is known about the mechanism of balance recovery in this group and what comprises a stable response to perturbation when simulating falls. There is also little understanding of the influences of pain and executive function on biomechanical responses to perturbation in this group. Importantly, there has been no investigation, to date, for predicting falls in older adults with knee OA using biomechanical parameters related to balance. The aims of this thesis were to investigate (1) the differences in balance response during induced falls in people with knee OA compared to asymptomatic controls in three trial conditions, (2) the relationship between pain and executive function on balance response, (3) the prediction of falls in older adults with knee OA using the biomechanical parameters associated with balance recovery. Forty-eight older people with knee OA (age average 71.02 ± 6.76 years, BMI average 29.10 ± 4.58 kg/m2, 54% females) and 15 asymptomatic older adults (age average 72.47 ± 4.81 years, BMI average 26.17 ± 3.06 kg/m2, 27% females) were recruited. Participants were placed in an induced lean position and were required to recovery balance when released. Spatio-temporal, upper, and lower limb kinematic and kinetic variables were analysed and compared between the two groups in three different trial types: no additional challenge (neither cognitive nor physical dual-task challenge), cognitive dual-task challenge, and physical dual-task challenge. A convenience sample of 24 OA participants was also selected to complete pain questionnaires and executive function assessments. Finally, participants in the convenience sample completed 12 months prospective falls calendars following baseline data collection. Based on reports of falls in this period, the sample was classified into fallers and non-fallers. The biomechanical measures during balance recovery and other data such as medication usage and patient demographics were used to fit a logistic regression model to predict fallers and non-fallers. When compared to controls, in all conditions, the OA group showed (1) slower and reduced spatio-temporal responses, (2) less ability to absorb impact forces at the knee, and (3) reduced ability to correct motion of the upper body posture. When compared to no additional challenge trials, in the dual-tasking trials there was (4) slower and smaller spatio-temporal responses and (5) greater knee power absorption, and (6) there was significantly reduced knee motion in the OA group. There was some correlation between unstable balance response and pain, in particular reduced hip flexion angular velocity and increased fear of severe pain (moderate positive, p = .02), and increased fear of total pain (moderate positive, p = .03). There was no correlation between unstable response and executive function. Finally, high centre of mass velocity and negative (extension) knee moment during balance recovery tasks were found to be good predictors of falls in older adults with knee OA. The relationship between extension knee moment and prediction of falls was stronger in women. The results from this thesis revealed that, compared to asymptomatic controls, older adults with knee OA took slower and shorter steps, and had a more upright posture following perturbation. Despite the deleterious influence of the spatio-temporal measures arising from the shorter and slower steps, this upright position of the trunk appears to play a part in increasing stability in the OA group via controlling motion of the two thirds of the body’s mass. Increased fear and interference from pain was correlated with lower hip flexion angle, which may also play a part in upright posture. Modelling results suggest a combination of upper body kinematics (velocity of centre of mass) and lower limb kinetics (knee extension moment) could be used to predict future falls in older women with knee OA.

The aim of this thesis was to increase our understanding of how virtual reality (VR) can be applied to assess and train pattern recognition and decision-making skill in sport, specifically the sport of tennis. There has been a growing interest in using VR for training perceptual- cognitive skill in sport; however, for VR training to effectively simulate real-world performance, it must recreate the contextual information and movement behaviours present in the real-world environment. Although it is well established that skilled performers can effectively use prior sources of contextual information to enhance anticipation performance compared to lesser skilled performers, little is known about the relative difficulty of identifying different types of contextual information and the requisite regularity of patterns to influence anticipation. Moreover, there is a lack of research assessing the effect of using more representative experimental tasks on anticipation and decision-making behaviour. Therefore, study one of this thesis assessed the representativeness of VR for simulation of tennis performance. Participants included 28 skilled tennis players aged between 12 to 17 years (M = 14.4, SD = 1.6). Participants sense of presence was assessed VR, and participants movement behaviours were compared when playing tennis in VR and real-world environments. The results showed that when performing groundstrokes, participants frequently used the same stance in VR as they did in the real-world condition and experienced a high sense of presence. Study two of this thesis used VR to assess the ability of 28 skilled tennis players aged between 13 and 18 years (M = 15.7, SD = 1.4) to identify two specific serving patterns being used by opponents. These serving patterns related to the opponent’s action tendencies, with a wide serve pattern connected to the side of the court the point started from (advantage side), and a tee serve pattern connected to the point score in the game (0-0). Participants were assessed on their ability to identify serving patterns by controlling how frequently patterns occurred during matches. Results revealed that patterns need to occur at high frequencies (100% of the time) during matches for skilled juniors to utilise this information to inform their anticipation responses. Study three of this thesis used VR to train 5 skilled tennis players aged between 14 and 18 years (M = 16, SD = 1.67) to utilise patterns of play when they occur at lower frequencies (80% of the time). Additionally, the influence of explicit instructions and no-instruction on learning and performance under pressure was assessed. It was found that exposure to patterns coupled with explicit instructions resulted in faster changes to response time and response accuracy performance, compared to no-instruction learning. Furthermore, instructions during training did not affect performance under pressure conditions. Overall, this thesis extends the perceptual-cognitive skill literature through its use of VR technology and methods of assessing task representativeness. Moreover, this thesis helps guide the design of future perceptual-cognitive skill research through the manipulation of contextual information in the VR environment and use of more implicit and explicit instructional methods to train decision-making performance.

Tennis coaches’ technical knowledge plays a critical part in performance analysis as it impacts their recommendations to athletes. To date, the research on tennis coaches’ knowledge is limited, and no previous studies have attempted to examine expert tennis coaches’ knowledge and diagnostic ability of the tennis serve technique. This research aimed to examine the knowledge and diagnostic ability of expert and novice tennis coaches, and to identify the key distinguishing characteristics between expert and novice tennis coaches. As a result, the key distinguishing characteristics between expert and novice tennis coaches were identified, and there was a recognised need to develop an online training course to improve coaches’ knowledge and understanding of the tennis serve technique. An online coach education course was then developed, and the effectiveness of the course was evaluated. This thesis comprised three studies employing a mixed-methods design. Data was collected and analysed by using both qualitative and quantitative methods to gain a deeper understanding of novice and expert tennis coaches’ knowledge, diagnostic skills and learning. The main data collection methods were in-depth interview and an online questionnaire that was constructed specifically for this investigation. Pre- and post-intervention design was used to investigate the efficacy of a training intervention on coaches’ knowledge and their diagnostic ability of the tennis serve technique. A video-based test was developed to measure the effectiveness of the online course. As a result of the first two studies, models representing declarative and practical knowledge of expert and novice tennis coaches were developed. The key distinguishing characteristics between novice and expert coaches were that experts displayed a superior understating of the functionality of the tennis serve technique and an advanced diagnostic ability. Expert coaches were able to see the tennis serve as “a whole” in comparison with novices who focused more on separate technical elements. It was suggested that novice coaches need to develop the knowledge of tennis serve from two perspectives: technical and functional. The findings of this thesis extend our current understanding of tennis coaches’ knowledge and diagnostic skills. The development and evaluation of an online training course provides a unique contribution to coach education in order to understand the cause-and-effect relationships between technical elements withing the flat serve should allow tennis coaches to prescribe more effective training interventions. By knowing how one element may affect another, coaches will be able to identify and rectify technical issues by addressing the cause of the problem instead of focusing on many separate elements. The design of this program can also be used with other technical elements in tennis and can be applied to other sports to improve coaches’ knowledge and diagnostic ability of sport technique.

Polycystic ovary syndrome (PCOS) is a common and complex endocrinopathy with significant metabolic and reproductive manifestations, carrying a major health and economic burden. Exercise has consistently been found to improve clinical outcomes, but shortfalls with exercise prescription are evident, with little known about the impact of exercise intensity for improving health outcomes. Research suggests that high-intensity interval training (HIIT) is feasible, well tolerated and enjoyable for people with, or at risk of, chronic disease and can address many of the shortfalls and barriers to exercise participation. Despite this, there is limited research exploring the efficacy of HIIT in comparison to standard moderate-intensity exercise recommendations for improving the health of women with PCOS. Therefore, the optimal exercise intensity for improving health outcomes remains unknown. The overall aim of this thesis was to determine the effectiveness of exercise, and more specifically exercise intensity, on the metabolic, reproductive and mental health and health-related quality of life of women with PCOS. The first two studies of my thesis were systematic reviews (Chapters 2 and 3), one that included a meta- analysis, to summarise and evaluate the current exercise intervention literature on key clinical metabolic, reproductive and anthropometric outcomes and on mental health and health-related quality of life (HRQoL) in women with PCOS. Briefly, results from these reviews suggested that exercise is beneficial for improving both the physical and mental health of women with PCOS. Furthermore, exercise of vigorous intensities provided additional benefits for increasing cardiorespiratory fitness, body composition and insulin resistance. There was insufficient evidence to be able to analyse the impact of exercise intensity on mental health and HRQoL, however, the available evidence indicated that exercise is effective for improving health-related quality of life and PCOS symptom distress. Exercise also shows some efficacy for improving symptoms and/or prevalence of depression and anxiety in women with PCOS. Following this, a protocol for a clinical trial was designed (Chapter 4), and conducted to investigate whether HIIT could confer greater benefits than standard moderate intensity continuous training (MICT) on cardio-metabolic outcomes (Chapter 5), reproductive outcomes (Chapter 6) and mental health and HRQoL (Chapter 7) in overweight women with PCOS. To do this, we employed a two-arm, randomised clinical trial where twenty-four overweight women, aged 18-45 years with diagnosed PCOS were randomised to either MICT (60-75% peak HR [HRpeak]) or HIIT (>90% HRpeak). We had 13 women complete the HIIT intervention and 11 women complete the MICT intervention. In regards to cardio-metabolic outcomes, both HIIT and MICT improved VO2peak (HIIT; 23.4 ± 10.1%, P

An increasingly useful application of machine learning (ML) is in predicting features of human actions. If it can be shown that algorithm inputs related to actual movement mechanics can predict a limb or limb segment’s future trajectory, a range of apparently intractable problems in movement science could be solved. The forecasting of lower limb trajectories can anticipate movement characteristics that may predict the risk of tripping, slipping or balance loss. Particularly in the design of human augmentation technology such as the exoskeleton, human movement prediction will improve the synchronisation between the user and the device greatly enhancing its efficacy. Long Short Term Memory (LSTM) neural neworks are a subset of ML algoithms that proven a wide success in modelling the human movement data. The aim of this thesis was to examine four LSTM neural nework architectures (Vanilla, Stacked, Bidirectional and Autoencoders) in predicting the future trajectories of lower limb kinematics, i.e. Angular Velocity (AV) and Linear Acceleration (LA). This work also aims to investigate whether linear statistical methods such as the Linear Regression (LR) is enough to predict the trajectories of lower limb kinematics. Kinematics data (LA and AV) of foot, shank and thigh were collected from 13 male and 3 female participants (28 ± 4 years old, 1.72 ± 0.07 m in height, 66 ± 10 kg in mass) who walked for 10 minutes at 4 different walking speeds on a 0% gradient treadmill. Walking -1 -1 speeds included preferred walking speed (PWS 4.34 ± 0.43 km.h ), imposed speed (5km.h , 15.4% ± 7.6% faster), slower speed (-20% PWS 3.59 ± 0.47 km.h-1) and faster speed (+20% PWS 5.26 ± 0.53 km.h-1). The sliding window technique was adopted for training and testing the LSTM models with total kinematics time-series data of 17,638 strides for all trials. The aim and findings of this work were carried out in 3 studies. Study 1 confirmed the possibility of predicting the future trajectories of human lower limb kinematics using LSTM autoencoders (ED-LSTM) and the LR during an imposed walking speed (5km.h-1). Both models achieved satisfactory predicted trajectories up to 0.06s. A prediction horizon of 0.06s can be used to compensate for delays in an exoskeleton’s feed-forward controller to better estimate the human motions and synchronise with intended movement trajectories. Study 2 (Chapter 4) indicated that the LR model is not suitable for the prediction of future lower limb kinematics at PWS. The LSTM perfromace results suggested that the ED-LSTM and the Stacked LSTM are more accurate to predict the future lower limb kinematics up to 0.1s at PWS and imposed walking speed (5km.h-1). The average duration for a gait cycle rages between 0.98-1.07s, and a prediction horizon of 0.1 accounts for about 10% of the gait cycle. Such a forecast may assist users in anticipating a low foot clearance to develop early countermeasures such as slowing down or stopping. Study 3 (Chapter 5) have shown that at +20% PWS the LSTM models’ performance obtained better predictions compared to all tested walking speed conditions (i.e. PWS, -20% PWS and 5km.h-1). While at -20% PWS, results indicated that at slower walking speeds all of the LSTM architectures obtained weaker predictions compared to all tested walking speeds (i.e. PWS, +20% PWS and 5km.h-1). In addition to the applications of a known future trajectories at the PWS mentioned in study 1 and 2, the prediction at fast and slow walking speeds familiarise the developed ML models with changes in human walking speed which are known to have large effects on lower limb kinematics. When intelligent ML methods are familiarised with the degree of kinematic changes due to speed variations, it could be used to improve human-machine interface in bionics design for various walking speeds The key finding of the three studies is that the ED-LSTM was found to be the most accurate -1 model to predict and adapt to the human motion kinematics at PWS, ±20% PWS and 5km.h up to 0.1s. The ability to predict future lower limb motions may have a wide range of applications including the design and control of bionics allowing better human-machine interface and mitigating the risk of tripping and balance loss.