Search

Abstract Long extrachromosomal circular DNA (leccDNA) regulates several biological processes such as genomic instability, gene amplification, and oncogenesis. The identification of leccDNA holds significant importance to investigate its potential associations with cancer, autoimmune, cardiovascular, and neurological diseases. In addition, understanding these associations can provide valuable insights about disease mechanisms and potential therapeutic approaches. Conventionally, wet lab-based methods are utilized to identify leccDNA, which are hindered by the need for prior knowledge, and resource-intensive processes, potentially limiting their broader applicability. To empower the process of leccDNA identification across multiple species, the paper in hand presents the very first computational predictor. The proposed iLEC-DNA predictor makes use of SVM classifier along with sequence-derived nucleotide distribution patterns and physicochemical properties-based features. In addition, the study introduces a set of 12 benchmark leccDNA datasets related to three species, namely Homo sapiens (HM), Arabidopsis Thaliana (AT), and Saccharomyces cerevisiae (SC/YS). It performs large-scale experimentation across 12 benchmark datasets under different experimental settings using the proposed predictor, more than 140 baseline predictors, and 858 encoder ensembles. The proposed predictor outperforms baseline predictors and encoder ensembles across diverse leccDNA datasets by producing average performance values of 81.09%, 62.2% and 81.08% in terms of ACC, MCC and AUC-ROC across all the datasets. The source code of the proposed and baseline predictors is available at https://github.com/FAhtisham/Extrachrosmosomal-DNA-Prediction . To facilitate the scientific community, a web application for leccDNA identification is available at https://sds_genetic_analysis.opendfki.de/iLEC_DNA/.

Survival prediction integrates patient-specific molecular information and clinical signatures to forecast the anticipated time of an event, such as recurrence, death, or disease progression. Survival prediction proves valuable in guiding treatment decisions, optimizing resource allocation, and interventions of precision medicine. The wide range of diseases, the existence of various variants within the same disease, and the reliance on available data necessitate disease-specific computational survival predictors. The widespread adoption of artificial intelligence (AI) methods in crafting survival predictors has undoubtedly revolutionized this field. However, the ever-increasing demand for more sophisticated and effective prediction models necessitates the continued creation of innovative advancements. To catalyze these advancements, it is crucial to bring existing survival predictors knowledge and insights into a centralized platform. The paper in hand thoroughly examines 23 existing review studies and provides a concise overview of their scope and limitations. Focusing on a comprehensive set of 90 most recent survival predictors across 44 diverse diseases, it delves into insights of diverse types of methods that are used in the development of disease-specific predictors. This exhaustive analysis encompasses the utilized data modalities along with a detailed analysis of subsets of clinical features, feature engineering methods, and the specific statistical, machine or deep learning approaches that have been employed. It also provides insights about survival prediction data sources, open-source predictors, and survival prediction frameworks.

Since the advent of deep learning (DL), the field has witnessed a continuous stream of innovations. However, the translation of these advancements into practical applications has not kept pace, particularly in safety-critical domains where artificial intelligence (AI) must meet stringent regulatory and ethical standards. This is underscored by the ongoing research in eXplainable AI (XAI) and privacy-preserving machine learning (PPML), which seek to address some limitations associated with these opaque and data-intensive models. Despite brisk research activity in both fields, little attention has been paid to their interaction. This work is the first to thoroughly investigate the effects of privacy-preserving techniques on explanations generated by common XAI methods for DL models. A detailed experimental analysis is conducted to quantify the impact of private training on the explanations provided by DL models, applied to six image datasets and five time series datasets across various domains. The analysis comprises three privacy techniques, nine XAI methods, and seven model architectures. The findings suggest non-negligible changes in explanations through the implementation of privacy measures. Apart from reporting individual effects of PPML on XAI, the paper gives clear recommendations for the choice of techniques in real applications. By unveiling the interdependencies of these pivotal technologies, this research marks an initial step toward resolving the challenges that hinder the deployment of AI in safety-critical settings.

Deep neural networks have demonstrated exceptional performance breakthroughs in the field of document image classification; yet, there has been limited research in the field that delves into the explainability of these models. In this paper, we present a comprehensive study in which we analyze 9 different explainability methods across 10 different state-of-the-art document classification models and 2 popular benchmark datasets, RVL-CDIP and Tobacco3482, making three major contributions. First, through an exhaustive qualitative and quantitative analysis of various explainability approaches, we demonstrate that the majority of them perform poorly in generating useful explanations for document images. Only two techniques, Occlusion and DeepSHAP, provide relatively faithful explanations, with DeepSHAP additionally offering better interpretability while Occlusion falls short in this regard. Second, to pinpoint the features most crucial to the models’ predictions, we present an approach for generating counterfactual explanations, the analysis of which reveals that many document classification models can be highly susceptible to minor perturbations in the input. Additionally, it suggests that these models may easily fall victim to biases in the document data, ultimately relying on seemingly irrelevant features to make their decisions. Specifically, on the RVL-CDIP dataset, we show that 25-50% of the overall model predictions and up to 60% of predictions for some classes were strongly dependent on these irrelevant features. Lastly, our analysis reveals that the popular document benchmark datasets, RVL-CDIP and Tobacco3482, are inherently biased, with document identification (ID) numbers of specific styles consistently appearing in certain document regions. If unaddressed, this bias allows the models to predict document classes solely by looking at the ID numbers and prevents them from learning more complex document features. Overall, by unveiling the strengths and weaknesses of various explainability methods, document datasets and deep learning models, our work presents a major step towards creating more transparent and robust AI-powered document image classification systems.

Summary: Environmental enteric dysfunction (EED) is a subclinical enteropathy challenging to diagnose due to an overlap of tissue features with other inflammatory enteropathies. EED subjects (n = 52) from Pakistan, controls (n = 25), and a validation EED cohort (n = 30) from Zambia were used to develop a machine-learning-based image analysis classification model. We extracted histologic feature representations from the Pakistan EED model and correlated them to transcriptomics and clinical biomarkers. In-silico metabolic network modeling was used to characterize alterations in metabolic flux between EED and controls and validated using untargeted lipidomics. Genes encoding beta-ureidopropionase, CYP4F3, and epoxide hydrolase 1 correlated to numerous tissue feature representations. Fatty acid and glycerophospholipid metabolism-related reactions showed altered flux. Increased phosphatidylcholine, lysophosphatidylcholine (LPC), and ether-linked LPCs, and decreased ester-linked LPCs were observed in the duodenal lipidome of Pakistan EED subjects, while plasma levels of glycine-conjugated bile acids were significantly increased. Together, these findings elucidate a multi-omic signature of EED.

ABSTRACT Environmental enteric dysfunction (EED) is a subclinical syndrome of altered small intestinal function postulated to be an important contributor to childhood undernutrition. The role of small intestinal bacterial communities in the pathophysiology of EED is poorly defined due to a paucity of studies where there has been a direct collection of small intestinal samples from undernourished children. Sixty-three members of a Pakistani cohort identified as being acutely malnourished between 3 and 6 months of age and whose wasting (weight-for-length Z-score [WLZ]) failed to improve after a 2-month nutritional intervention underwent esophagogastroduodenoscopy (EGD). Paired duodenal luminal aspirates and duodenal mucosal biopsies were obtained from 43 children. Duodenal microbiota composition was characterized by sequencing bacterial 16S rRNA gene amplicons. Levels of bacterial taxa (amplicon sequence variants [ASVs]) were referenced to anthropometric indices, histopathologic severity in biopsies, expression of selected genes in the duodenal mucosa, and fecal levels of an immunoinflammatory biomarker (lipocalin-2). A “core” group of eight bacterial ASVs was present in the duodenal samples of 69% of participants. Streptococcus anginosus was the most prevalent, followed by Streptococcus sp., Gemella haemolysans, Streptococcus australis, Granulicatella elegans, Granulicatella adiacens, and Abiotrophia defectiva. At the time of EGD, none of the core taxa were significantly correlated with WLZ. Statistically significant correlations were documented between the abundances of Granulicatella elegans and Granulicatella adiacens and the expression of duodenal mucosal genes involved in immune responses (dual oxidase maturation factor 2, serum amyloid A, and granzyme H). These results suggest that a potential role for members of the oral microbiota in pathogenesis, notably Streptococcus, Gemella, and Granulicatella species, warrants further investigation.IMPORTANCEUndernutrition among women and children is a pressing global health problem. Environmental enteric dysfunction (EED) is a disease of the small intestine (SI) associated with impaired gut mucosal barrier function and reduced capacity for nutrient absorption. The cause of EED is ill-defined. One emerging hypothesis is that alterations in the SI microbiota contribute to EED. We performed a culture-independent analysis of the SI microbiota of a cohort of Pakistani children with undernutrition who had failed a standard nutritional intervention, underwent upper gastrointestinal tract endoscopy, and had histologic evidence of EED in their duodenal mucosal biopsies. The results revealed a shared group of bacterial taxa in their duodenums whose absolute abundances were correlated with levels of the expression of genes in the duodenal mucosa that are involved in inflammatory responses. A number of these bacterial taxa are more typically found in the oral microbiota, a finding that has potential physiologic and therapeutic implications.

Abstract Niacin had long been understood as an antioxidant. There were reports that high fat diet (HFD) may cause psychological and physical impairments. The present study was aimed to experience the effect of Niacin on % growth rate, cumulative food intake, motor activity and anxiety profile, redox status, 5-HT metabolism and brain histopathology in rats. Rats were administered with Niacin at a dose of 50 mg/ml/kg body weight for 4 weeks following normal diet (ND) and HFD. Behavioral tests were performed after 4 weeks. Animals were sacrificed to collect brain samples. Biochemical, neurochemical and histopathological studies were performed. HFD increased food intake and body weight. The exploratory activity was reduced and anxiety like behavior was observed in HFD treated animals. Activity of antioxidant enzymes was decreased while oxidative stress marker and serotonin metabolism in the brain of rat were increased in HFD treated animals than ND fed rats. Morphology of the brain was also altered by HFD administration. Conversely, Niacin treated animals decreased food intake and % growth rate, increased exploratory activity, produced anxiolytic effects, decreased oxidative stress and increased antioxidant enzyme and 5-HT levels following HFD. Morphology of brain is also normalized by the treatment of Niacin following HFD. In-silico studies showed that Niacin has a potential binding affinity with degradative enzyme of 5-HT i.e. monoamine oxidase (MAO) A and B with an energy of ~ − 4.5 and − 5.0 kcal/mol respectively. In conclusion, the present study showed that Niacin enhanced motor activity, produced anxiolytic effect, and reduced oxidative stress, appetite, growth rate, increased antioxidant enzymes and normalized serotonin system and brain morphology following HFD intake. In-silico studies suggested that increase 5-HT was associated with the binding of MAO with Niacin subsequentially an inhibition of the degradation of monoamine. It is suggested that Niacin has a great antioxidant potential and could be a good therapy for the treatment of HFD induced obesity.

Abstract The acquisition of antimicrobial resistance (AR) genes has rendered important pathogens nearly or fully unresponsive to antibiotics. It has been suggested that pathogens acquire AR traits from the gut microbiota, which collectively serve as a global reservoir for AR genes conferring resistance to all classes of antibiotics. However, only a subset of AR genes confers resistance to clinically relevant antibiotics, and, although these AR gene profiles are well-characterized for common pathogens, less is known about their taxonomic associations and transfer potential within diverse members of the gut microbiota. We examined a collection of 14,850 human metagenomes and 1666 environmental metagenomes from 33 countries, in addition to nearly 600,000 isolate genomes, to gain insight into the global prevalence and taxonomic range of clinically relevant AR genes. We find that several of the most concerning AR genes, such as those encoding the cephalosporinase CTX-M and carbapenemases KPC, IMP, NDM, and VIM, remain taxonomically restricted to Proteobacteria. Even cfiA, the most common carbapenemase gene within the human gut microbiome, remains tightly restricted to Bacteroides, despite being found on a mobilizable plasmid. We confirmed these findings in gut microbiome samples from India, Honduras, Pakistan, and Vietnam, using a high-sensitivity single-cell fusion PCR approach. Focusing on a set of genes encoding carbapenemases and cephalosporinases, thus far restricted to Bacteroides species, we find that few mutations are required for efficacy in a different phylum, raising the question of why these genes have not spread more widely. Overall, these data suggest that globally prevalent, clinically relevant AR genes have not yet established themselves across diverse commensal gut microbiota.

In this study, the resistance of selected canola genotypes to the mustard aphid Lipaphis erysimi (Kaltenbach) was investigated. Molecular characterization of plant resistance was conducted, and genetic variability was observed between the resistant (KS-75) and susceptible (Abaseen) genotypes of brassica. Three RAPD markers were identified that were associated with the resistance. This study was conducted to compare the genetic data between two genotypes (KS-75 resistant and Abaseen susceptible), with an average of 4.3 bands per primer. The amplification bands per primer ranged from 9 to polymorphic fragments of 8, with fragment sizes ranging from 450 to 1700 base pairs to 9 kilo base pairs (kbps). Amplification of the largest fragment in 9 kbps using four primers (A 1–12, B 1–12, K 1–12 and L 1–12) resulted in a mean 92 percent identity index between the first two resistant and susceptible genotypes. KS-75 resistant had the highest mean genetic diversity of 92 percent, while Abaseen susceptible had the lowest mean genetic diversity of 8 percent. Out of the three markers tested, two were found to be positively associated with aphid resistance, while one was negatively correlated. These markers can be used to help develop a pre-emptive strategy for aphid resistance in brassica species, without having to simultaneously select for yield and quality related traits.

Abstract The increasing complexity of low‐voltage networks poses a growing challenge for the reliable and fail‐safe operation of electricity grids. The reasons for this include an increasingly decentralised energy generation (photovoltaic systems, wind power etc.) and the emergence of new types of consumers (e‐mobility, domestic electricity storage etc.). At the same time, the low‐voltage grid is largely unmonitored and local power failures are sometimes hard to detect. To overcome this, power line communication (PLC) has emerged as a potential solution for reliable monitoring of the low‐voltage grid. In addition to establishing a communication infrastructure, PLC also offers the possibility of evaluating the cables themselves, as well as the connection quality between individual cable distributors based on their signal‐to‐noise ratio (SNR). The roll‐out of a large‐scale PLC infrastructure therefore not only ensures communication, but also introduces a tool for monitoring the entire network. To evaluate the potential of this data, we installed 38 PLC modems in three different areas of a German city with a population of about 150,000 as part of the Fühler‐im‐Netz (FiN) project. Over a period of 22 months, an SNR spectrum of each connection between adjacent PLC modems was generated every quarter of an hour. The availability of this real‐world PLC data opens up new possibilities to react to the increasingly complex challenges in future smart grids. This paper provides a detailed analysis of the data generation and describes how the data was collected during normal operation of the electricity grid. In addition, we present common anomalies, effects, and trends that could be observed in the PLC data at daily, weekly, or seasonal levels. Finally, we discuss potential use cases and the remote inspection of a cable section is highlighted as an example.

Purpose: The study determines employee perception of risk management practices and organizational commitment in Private Banks of Balochistan. Earlier studies in the field highlight the importance of HR practices during organizational transition since it may be necessary to influence employee behavior to support the change's direction. By implementing HR practices that encourage change, The HR department is in an excellent position to effect positive perception. Methodology: Using a quantitative research approach, this study evaluates the relationship between employee perception of risk management practices and organizational commitment in the banking sector. Findings: The study's findings for banking sector organizations, supporting them in better understanding the elements that influence employee perceptions of risk management practices and their commitment to the organization. The study's findings could be used to design focused initiatives to improve risk management practices and increase employee happiness and organizational effectiveness. Implications: This study may provide a foundation for future research in related domains by contributing to existing knowledge on risk management and organizational commitment. Finally, the study hopes to promote a risk-aware culture within the banking sector, resulting in a more engaged and resilient workforce and boosting these organizations' long-term prosperity and stability.

Artificial Intelligence (AI) has achieved remarkable success in image generation, image analysis, and language modeling, making data-driven techniques increasingly relevant in practical real-world applications, promising enhanced creativity and efficiency for human users. However, the deployment of AI in high-stakes domains such as infrastructure and healthcare still raises concerns regarding algorithm accountability and safety. The emerging field of explainable AI (XAI) has made significant strides in developing interfaces that enable humans to comprehend the decisions made by data-driven models. Among these approaches, concept-based explainability stands out due to its ability to align explanations with high-level concepts familiar to users. Nonetheless, early research in adversarial machine learning has unveiled that exposing model explanations can render victim models more susceptible to attacks. This is the first study to investigate and compare the impact of concept-based explanations on the privacy of Deep Learning based AI models in the context of biomedical image analysis. An extensive privacy benchmark is conducted on three different state-of-the-art model architectures (ResNet50, NFNet, ConvNeXt) trained on two biomedical (ISIC and EyePACS) and one synthetic dataset (SCDB). The success of membership inference attacks while exposing varying degrees of attribution-based and concept-based explanations is systematically compared. The findings indicate that, in theory, concept-based explanations can potentially increase the vulnerability of a private AI system by up to 16% compared to attributions in the baseline setting. However, it is demonstrated that, in more realistic attack scenarios, the threat posed by explanations is negligible in practice. Furthermore, actionable recommendations are provided to ensure the safe deployment of concept-based XAI systems. In addition, the impact of differential privacy (DP) on the quality of concept-based explanations is explored, revealing that while negatively influencing the explanation ability, DP can have an adverse effect on the models’ privacy.

Abstract Background Exclusive breastfeeding (EBF) of children until six months of age is considered one of the most critical interventions in tackling childhood undernutrition. EBF rates are suboptimal in Pakistan, particularly in rural areas where child undernutrition is most prevalent. This study aimed to explore barriers to EBF in a rural context of Pakistan. Methods The study was conducted in the rural district Matiari of Sindh, Pakistan, during Jan-March 2020. We used a qualitative exploratory study design and conducted 36 focus group discussions (FGDs). Participants were purposively selected mothers who had not practiced EBF during their previous childbirth, their spouses and mothers-in-law, and lady health workers (LHWs) serving in the study catchment. FGDs were audio-recorded, transcribed, and translated into English from the local language and analysed using thematic content analysis. Results Barriers to EBF included low awareness and cultural practices of prelacteal feeds, insufficient breast milk production, undernutrition of mothers, mothers’ occupation as fieldworkers, less birth spacing, low awareness about the correct technique of breastfeeding, maternal and child ailments, abnormal breasts, and influence of in-laws to start top-up feeds. Several facilitators were identified: family support, appropriate maternal diet, maternal awareness, and support in the neighborhood. Conclusion Barriers to EBF are multifaceted in rural areas, and interventions aiming to improve adherence to EBFshould be multipronged. Awareness-raising alone might not be sufficient, and other interventions should be designed to address the barriers of maternal malnutrition, insufficient milk production, and socio-cultural practices. In addition, safe alternatives to breast milk may be necessary if breastfeeding is truly not feasible. Lack of breast milk substitutes is particularly challenging for poor rural women who cannot afford infant formula milk.

Abstract Deep exploration of histone occupancy and covalent post-translational modifications (e.g., acetylation, methylation) is essential to decode gene expression regulation, chromosome packaging, DNA damage, and transcriptional activation. Existing computational approaches are unable to precisely predict histone occupancy and modifications mainly due to the use of sub-optimal statistical representation of histone sequences. For the establishment of an improved histone occupancy and modification landscape for multiple histone markers, the paper in hand presents an end-to-end computational multi-paradigm framework “Histone-Net”. To learn local and global residue context aware sequence representation, Histone-Net generates unsupervised higher order residue embeddings (DNA2Vec) and presents a different application of language modelling, where it encapsulates histone occupancy and modification information while generating higher order residue embeddings (SuperDNA2Vec) in a supervised manner. We perform an intrinsic and extrinsic evaluation of both presented distributed representation learning schemes. A comprehensive empirical evaluation of Histone-Net over ten benchmark histone markers data sets for three different histone sequence analysis tasks indicates that SuperDNA2Vec sequence representation and softmax classifier-based approach outperforms state-of-the-art approach by an average accuracy of 7%. To eliminate the overhead of training separate binary classifiers for all ten histone markers, Histone-Net is evaluated in multi-label classification paradigm, where it produces decent performance for simultaneous prediction of histone occupancy, acetylation, and methylation.

Graphene is an advanced material in the carbon group and offers greater mechanical, electrical, structural, and optical properties. Graphene oxide (GO) and reduced graphene oxide (rGO) nanoparticles were synthesized and characterized and their special effects on enhancing the physio-mechanical characteristics of medium density fiberboard (MDF) were assessed. GO and rGO nanoparticles were added to urea formaldehyde (UF) resin at different weight percentages (1.0, 2.0, and 3.0 wt%) during the dosing process. To manufacture the MDF, nanofillers were created by sonication and combination with natural wood fibers. To observe the behavior of nanoparticles in the nanofillers, microstructure characterizations were conducted. The manufactured nano MDF samples underwent physical and mechanical testing. The incorporation of GO and rGO nanoparticles into UF resin led to significant improvements in the physical and mechanical properties of the MDF. The addition of GO and rGO nanoparticles at different weight percentages (1.0, 2.0, and 3.0 wt%) resulted in a range of improvements in thickness swelling (up to 53.3% and 35.2% for GO and rGO nanoparticles, respectively), water absorption (up to 23.3% and 63.15%, respectively), and thermal conductivity (up to 42.16% and 27.7%, respectively). Additionally, the internal bond and rupture modulus of the MDF was enhanced by 59.0% and 70.0%, respectively, for GO and 41.4% and 48.5% for rGO. The highest value of the modulus of rupture (MoR) was observed at a concentration of 3.0% of rGO nanoparticles (44.7 MPa). The findings also showed that thickness swelling (Ts) and water absorption (WA) exhibited directly proportional relationships for 3.0% GO and rGO. These results suggested that incorporating GO and rGO nanoparticles into UF resin can significantly improve the physical and mechanical properties of nano MDF.

This study assessed the impact of gross domestic product (GDP), education, natural resources, remittances, and financial inclusion on carbon emissions in G-11 countries from 1990 to 2021. Based on the negative impact of pollution and the need for sustainable development, this study examined factors affecting CO2 emissions in G-11 countries using non-linear panel ARDL model. The study found that a positive GDP shock increases CO2 emissions in the short and long term, while a negative shock decreases emissions in the short term and increases emissions in the long term. Education was found to increase CO2 emissions in the long term but decrease them in the short term, emphasizing the need for education on combating emissions. Natural resources were also found to increase emissions in the long term, highlighting the need for government-defined institutions to minimize extraction effects and enforce transparency and accountability. Positive changes in personal remittances and financial inclusion were found to increase emissions in both the short and long term, suggesting the need for policies that encourage renewable energy sources and energy efficiency improvement. The study concludes that policymakers should prioritize efficient resource allocation, promote renewable energy usage, and enhance environmental awareness to achieve sustainable development goals in G-11 countries. The possible applications of this study include the use of the models to investigate the asymmetric effects on CO2 emissions. This model can be applied in future studies to examine the relationship between GDP, education, natural resources, personal remittances, financial inclusion, and CO2 emissions in other countries.

Objectives: Staphylococcus aureus and Streptococcus pneumoniae are common colonizers of the human nasopharynx. In this study, we describe S. aureus nasopharyngeal carriage and evaluate its association with S. pneumoniae carriage post-10-valent pneumococcal conjugate vaccine (PCV10) introduction in Pakistan. Methods: A serial cross-sectional study was undertaken from 2014 to 2018, children

Subcellular localization of Ribonucleic Acid (RNA) molecules provide significant insights into the functionality of RNAs and helps to explore their association with various diseases. Predominantly developed single-compartment localization predictors (SCLPs) lack to demystify RNA association with diverse biochemical and pathological processes mainly happen through RNA co-localization in multiple compartments. Limited multi-compartment localization predictors (MCLPs) manage to produce decent performance only for target RNA class of particular sub-type. Further, existing computational approaches have limited practical significance and potential to optimize therapeutics due to the poor degree of model explainability. The paper in hand presents an explainable Long Short-Term Memory (LSTM) network “EL-RMLocNet”, predictive performance and interpretability of which are optimized using a novel GeneticSeq2Vec statistical representation learning scheme and attention mechanism for accurate multi-compartment localization prediction of different RNAs solely using raw RNA sequences. GeneticSeq2Vec generates optimized statistical vectors of raw RNA sequences by capturing short and long range relations of nucleotide k-mers. Using sequence vectors generated by GeneticSeq2Vec scheme, Long Short Term Memory layers extract most informative features, weighting of which on the basis of discriminative potential for accurate multi-compartment localization prediction is performed using attention layer. Through reverse engineering, weights of statistical feature space are mapped to nucleotide k-mers patterns to make multi-compartment localization prediction decision making transparent and explainable for different RNA classes and species. Empirical evaluation indicates that EL-RMLocNet outperforms state-of-the-art predictor for subcellular localization prediction of 4 different RNA classes by an average accuracy figure of 8% for Homo Sapiens species and 6% for Mus Musculus species. EL-RMLocNet is freely available as a web server at (https://sds_genetic_analysis.opendfki.de/subcellular_loc/).

Health mention classification (HMC) involves the classification of an input text as health mention or not. Figurative and non-health mention of disease words makes the classification task challenging. Learning the context of the input text is the key to this problem. The idea is to learn word representation by its surrounding words and utilize emojis in the text to help improve the classification results. In this paper, we improve the word representation of the input text using adversarial training that acts as a regularizer during fine-tuning of the model. We generate adversarial examples by perturbing the word embeddings of the model and then train the model on a pair of clean and adversarial examples. Additionally, we utilize contrastive loss that tries to learn similar representations for the clean example and its perturbed version. We train and evaluate the method on three public datasets. Experiments show that contrastive adversarial training improves the performance significantly in terms of F1-score over the baseline methods of both BERTLarge and RoBERTaLarge on all three datasets. Furthermore, we provide a brief analysis of the results by utilizing the power of explainable AI.

BackgroundThe reduction in severe and moderate acute malnutrition (SAM and MAM) rates in Pakistan has been sub-optimal compared to other low-and middle-income countries (LMICs). Specially-formulated products have been designed globally to manage SAM and MAM, such as ready-to-use therapeutic food (RUTF) and ready-to-use supplementary food (RUSF), with variable efficacies. RUTF is primarily produced and patented in industrialized countries, raising supply challenges in resource-constrained regions with a high burden of acute malnutrition. RUSF minimizes costs by using locally-available ingredients while providing similar nutritional value. In this study, we compared the efficacy, side effects, and compliance of two months of supplementation with either RUTF or RUSF.MethodsChildren aged nine months in the rural district of Matiari, Pakistan, with a weight-for-height z-score (WHZ) ResultsThe RUSF group had a higher height gain and mid-upper arm circumferences (MUAC) score. Higher compliance was noted with lower side effects in the RUSF group. A higher compliance rate did correlate with the growth parameters in respective groups.ConclusionOur study found that both RUTF and RUSF partially improve the anthropometric status of acutely malnourished children, with neither being superior to the other.

IntroductionEnvironmental enteropathy is an important contributor to childhood malnutrition in the developing world. Chronic exposure to fecal pathogens leads to alteration in intestinal structure and function, resulting in impaired gut immune function, malabsorption, and growth faltering leading to environmental enteropathy.MethodsA community-based intervention study was carried out on children till 24 months of age in Matiari district, Pakistan. Blood and fecal specimens were collected from the enrolled children aged 3–6 and 9 months. A real-time PCR-based TaqMan array card (TAC) was used to detect enteropathogens.ResultsGiardia, Campylobacter spp., enteroaggregative Escherichia coli (EAEC), Enteropathogenic Escherichia coli (EPEC), Enterotoxigenic Escherichia coli (ETEC), and Cryptosporidium spp. were the most prevailing enteropathogens in terms of overall positivity at both time points. Detection of protozoa at enrollment and 9 months was negatively correlated with rate of change in height-for-age Z (ΔHAZ) scores during the first and second years of life. A positive association was found between Giardia, fecal lipocalin (LCN), and alpha 1-Acid Glycoprotein (AGP), while Campylobacter spp. showed positive associations with neopterin (NEO) and myeloperoxidase (MPO).ConclusionProtozoal colonization is associated with a decline in linear growth velocity during the first 2 years of life in children living in Environmental enteric dysfunction (EED) endemic settings. Mechanistic studies exploring the role of cumulative microbial colonization, their adaptations to undernutrition, and their influence on gut homeostasis are required to understand symptomatic enteropathogen-induced growth faltering.

In Tharparkar desert, the main source of water supply is groundwater. Water for domestic use is normally collected from open dug wells. For this study, twenty-five groundwater samples were collected from the open dug well of Talukas of Tharparkar, Southern Sindh, Pakistan, to evaluate its quality and suitability for drinking, domestic, and agricultural use. The findings in this paper are based on an examination of regional variations and the use of a geographic information system (GIS), which incorporates the results of 25 well water samples collected from six villages in Taluka Mithi Tharparkar. The study's findings reveal excessive concentrations of physical and chemical characteristics in the majority of locations, in contradiction to the World Health Organization's (WHO) recommendations for drinking water. Majority of wells in the studied area the TDS were ranging from 800 to 11,000 mg/l. However, villages of Madhu, Dharar and Mithi were found to have high TDS that exceeded 1000 mg/L, which was a major cause of health problems in the area. Moreover, the total concentration of arsenic and fluoride in water samples varied between 0.0015 and 80 μg/L and 0.1–5.1 mg/l respectively. The inorganic form of As, was found to be 95% of the total As level in the regional drinking water. The concentration of Fluoride in areas of Mithi, Dharar, Harmar and Satar was found above the permissible limits. Furthermore, results indicate that correlations among water parameters dropped, and examined values were inconsequential. Statistical analysis (a linear regression method) was performed, for predicting the concentrations of groundwater quality parameters. A significant positive association between 1% and 4% was found between different parameters. The collected health data indicate that the study area has waterborne diseases. Therefore, comprehensive groundwater monitoring and management are required in the studied area to protect populations from various diseases caused by poor water quality.

Viral-host protein-protein interaction (VHPPI) prediction is essential to decoding molecular mechanisms of viral pathogens and host immunity processes that eventually help to control the propagation of viral diseases and to design optimized therapeutics. Multiple AI-based predictors have been developed to predict diverse VHPPIs across a wide range of viruses and hosts, however, these predictors produce better performance only for specific types of hosts and viruses. The prime objective of this research is to develop a robust meta predictor (MP-VHPPI) capable of more accurately predicting VHPPI across multiple hosts and viruses. The proposed meta predictor makes use of two well-known encoding methods Amphiphilic Pseudo-Amino Acid Composition (APAAC) and Quasi-sequence (QS) Order that capture amino acids sequence order and distributional information to most effectively generate the numerical representation of complete viral-host raw protein sequences. Feature agglomeration method is utilized to transform the original feature space into a more informative feature space. Random forest (RF) and Extra tree (ET) classifiers are trained on optimized feature space of both APAAC and QS order separate encoders and by combining both encodings. Further predictions of both classifiers are utilized to feed the Support Vector Machine (SVM) classifier that makes final predictions. The proposed meta predictor is evaluated over 7 different benchmark datasets, where it outperforms existing VHPPI predictors with an average performance of 3.07, 6.07, 2.95, and 2.85% in terms of accuracy, Mathews correlation coefficient, precision, and sensitivity, respectively. To facilitate the scientific community, the MP-VHPPI web server is available at https://sds_genetic_analysis.opendfki.de/MP-VHPPI/.

Summary: Protein-protein interaction (PPI) prediction is essential to understand the functions of proteins in various biological processes and their roles in the development, progression, and treatment of different diseases. To perform economical large-scale PPI analysis, several artificial intelligence-based approaches have been proposed. However, these approaches have limited predictive performance due to the use of in-effective statistical representation learning methods and predictors that lack the ability to extract comprehensive discriminative features. The paper in hand generates statistical representation of protein sequences by applying transfer learning in an unsupervised manner using FastText embedding generation approach. Furthermore, it presents “ADH-PPI” classifier which reaps the benefits of three different neural layers, long short-term memory, convolutional, and self-attention layers. Over two different species benchmark datasets, proposed ADH-PPI predictor outperforms existing approaches by an overall accuracy of 4%, and matthews correlation coefficient of 6%. In addition, it achieves an overall accuracy increment of 7% on four independent test sets. Availability: ADH-PPI web server is publicly available at https://sds_genetic_analysis.opendfki.de/PPI/

Small non-coding RNAs (ncRNAs) are attracting increasing attention as they are now considered potentially valuable resources in the development of new drugs intended to cure several human diseases. A prerequisite for the development of drugs targeting ncRNAs or the related pathways is the identification and correct classification of such ncRNAs. State-of-the-art small ncRNA classification methodologies use secondary structural features as input. However, such feature extraction approaches only take global characteristics into account and completely ignore co-relative effects of local structures. Furthermore, secondary structure based approaches incorporate high dimensional feature space which is computationally expensive. The present paper proposes a novel Robust and Precise ConvNet (RPC-snRC) methodology which classifies small ncRNAs into relevant families by utilizing their primary sequence. RPC-snRC methodology learns hierarchical representation of features by utilizing positioning and information on the occurrence of nucleotides. To avoid exploding and vanishing gradient problems, we use an approach similar to DenseNet in which gradient can flow straight from subsequent layers to previous layers. In order to assess the effectiveness of deeper architectures for small ncRNA classification, we also adapted two ResNet architectures having a different number of layers. Experimental results on a benchmark small ncRNA dataset show that the proposed methodology does not only outperform existing small ncRNA classification approaches with a significant performance margin of 10% but it also gives better results than adapted ResNet architectures. To reproduce the results Source code and data set is available at https://github.com/muas16/small-non-coding-RNA-classification.

Supervised learning, despite being extremely effective, relies on expensive, time-consuming, and error-prone annotations. Self-supervised learning has recently emerged as a strong alternate to supervised learning in a range of different domains as collecting a large amount of unlabeled data can be achieved by simply crawling the internet. These self-supervised methods automatically discover features relevant to represent an input example by using self-defined proxy tasks. In this paper, we question the potential of commonly employed purely supervised training (starting either from ImageNet pretrained networks or pure random initialization) in contrast to self-supervised representations that can be learned directly using self-supervised representation learning methods on large document image datasets. For this purpose, we leverage a large-scale document image collection (RVL-CDIP) to train ResNet-50 image encoder using two different self-supervision methods (SimCLR and Barlow Twins). Employing a linear classifier on top of self-supervised embeddings from ResNet-50 results in an accuracy of 86.75% as compared to 71.43% from the corresponding ImageNet pretrained embeddings. Similarly, evaluating on Tobacco-3482 dataset using self-supervised embeddings from ResNet-50 yields an accuracy of 88.52% in contrast to 74.16% from the corresponding ImageNet pretrained embeddings. We show that in the case of limited labeled data, this wide gap in performance between self-supervised and fully supervised models persists even after fine-tuning pretrained models. However, a significant reduction in this gap is observed with an increasing amount of data including the case where the model is trained from scratch. Our results show that representations learned using self-supervised representation learning techniques are a viable option for document image classification, specifically in the context of limited labeled data, which is a usual restriction in industrial use cases.

Abstract Background Stunting affects up to one-third of the children in low-to-middle income countries (LMICs) and has been correlated with decline in cognitive capacity and vaccine immunogenicity. Early identification of infants at risk is critical for early intervention and prevention of morbidity. The aim of this study was to investigate patterns of growth in infants up through 48 months of age to assess whether the growth of infants with stunting eventually improved as well as the potential predictors of growth. Methods Height-for-age z-scores (HAZ) of children from Matiari (rural site, Pakistan) at birth, 18 months, and 48 months were obtained. Results of serum-based biomarkers collected at 6 and 9 months were recorded. A descriptive analysis of the population was followed by assessment of growth predictors via traditional machine learning random forest models. Results Of the 107 children who were followed up till 48 months of age, 51% were stunted (HAZ

Traditional neural networks trained using point-based maximum likelihood estimation are deterministic models and have exhibited near-human performance in many image classification tasks. However, their insistence on representing network parameters with point-estimates renders them incapable of capturing all possible combinations of the weights; consequently, resulting in a biased predictor towards their initialisation. Most importantly, these deterministic networks are inherently unable to provide any uncertainty estimate for their prediction which is highly sought after in many critical application areas. On the other hand, Bayesian neural networks place a probability distribution on network weights and give a built-in regularisation effect making these models able to learn well from small datasets without overfitting. These networks provide a way of generating posterior distribution which can be used for model's uncertainty estimation. However, Bayesian estimation is computationally very expensive since it greatly widens the parameter space. This paper proposes a hybrid convolutional neural network which combines high accuracy of deterministic models with posterior distribution approximation of Bayesian neural networks. This hybrid architecture is validated on 13 publicly available benchmark classification datasets from a wide range of domains and different modalities like natural scene images, medical images, and time-series. Our results show that the proposed hybrid approach performs better than both deterministic and Bayesian methods in terms of classification accuracy and also provides an estimate of uncertainty for every prediction. We further employ this uncertainty to filter out unconfident predictions and achieve significant additional gain in accuracy for the remaining predictions.

The revolution of the Internet of Things (IoT) is transforming many concepts, making them “Smart”. It has revolutionized many areas of real life. Smart City is one of the key concepts of this revolution. Although cities are being digitally transformed, it still has obstacles along the way. In this paper, we have analyzed different publications for our Systematic Literature Review (SLR). This study highlights the areas where the blockchain is used and determines the benefits of using blockchain. The main contribution is to explore and identify the barriers and hurdles in Smart City Domain and how these hurdles are mitigated by the blockchain technology. This Systematic Study also addresses different challenges and issues such as security, immutability, interoperability, decentralization, privacy, and trust in the development of Smart Cities. An overview of the systematic research is also presented that would help identify the most and least studied concerns addressed in this study. This paper aims at analyzing how blockchain technology is used in different smart city business models, what are the features of blockchain that could consider it to be used beyond crypto currencies. We hope that this study can inspire interest in theory and exercise to stand-in discussions in this area adhere to these limits.

Publications are an integral part of a scientific community. Bibliographic reference extraction from scientific publication is a challenging task due to diversity in referencing styles and document layout. Existing methods perform sufficiently on one dataset however, applying these solutions to a different dataset proves to be challenging. Therefore, a generic solution was anticipated which could overcome the limitations of the previous approaches. The contribution of this paper is three-fold. First, it presents a novel approach called DeepBiRD which is inspired by human visual perception and exploits layout features to identify individual references in a scientific publication. Second, we release a large dataset for image-based reference detection with 2401 scans containing 38863 references, all manually annotated for individual reference. Third, we present a unified and highly configurable end-to-end automatic bibliographic reference extraction framework called BRExSys which employs DeepBiRD along with state-of-the-art text-based models to detect and visualize references from a bibliographic document. Our proposed approach pre-processes the images in which a hybrid representation is obtained by processing the given image using different computer vision techniques. Then, it performs layout driven reference detection using Mask R-CNN on a given scientific publication. DeepBiRD was evaluated on two different datasets to demonstrate the generalization of this approach. The proposed system achieved an AP50 of 98.56% on our dataset. DeepBiRD significantly outperformed the current state-of-the-art approach on their dataset. Therefore, suggesting that DeepBiRD is significantly superior in performance, generalized, and independent of any domain or referencing style.

In-air signature verification is vital for biometric user identification in contact-less mode. The state-of-the-art methods use heuristics for signature acquisition, and provide insufficient data to train neural networks for the verification. In this article, we present a novel method for end-to-end deep learning based in-air signature verification using a depth sensor. In this regard, we propose a new medium-scale in-air signature dataset which is created using an accurate convolutional neural network (CNN) based 3D hand pose estimation algorithm. The proposed dataset offers a total of 1800 signatures collected from 40 subjects. So far, dynamic time warping (DTW) has been the most effective and commonly used method for verification. Keeping in view the significant advancement in deep learning, we present a more accurate deep learning based alternative which outperforms DTW by 67.6%. To this end, we train a personalized autoencoder to reconstruct the signature of each subject. Thereafter, the signature is verified by thresholding the reconstruction loss. We perform extensive experiments by formulating spatial and depth features of the signature in images and point clouds based representations. Moreover, for comparisons, we implement several deep learning algorithms (i.e. linear autoencoder, convolutional autoencoder, and Deep One-Class classifier). Our verification approach achieves an EER of 0.055%. The dataset is available at https://bit.ly/2mEJzOw.

Viral-host protein protein interaction (PPI) analysis is essential to decode the molecular mechanism of viral pathogen and host immunity processes which eventually help to control viral diseases and optimize therapeutics. The state-of-the-art viral-host PPI predictor leverages unsupervised embedding learning technique (doc2vec) to generate statistical representations of viral-host protein sequences and a Random Forest classifier for interaction prediction. However, doc2vec approach generates the statistical representations of viral-host protein sequences by merely modelling the local context of residues which only partially captures residue semantics. The paper in hand proposes a novel technique for generating better statistical representations of viral and host protein sequences based on the infusion of comprehensive local and global contextual information of the residues. While local residue context aware encoding captures semantic relatedness and short range dependencies of residues. Global residue context aware encoding captures comprehensive long-range residues dependencies, positional invariance of residues, and unique residue combination distribution important for interaction prediction. Using concatenated rich statistical representations of viral and host protein sequences, a robust machine learning framework "LGCA-VHPPI" is developed which makes use of a deep forest model to effectively model complex non-linearity of viral-host PPI sequences. An in-depth performance comparison of the proposed LGCA-VHPPI framework with existing diverse sequence encoding schemes based viral-host PPI predictors reveals that LGCA-VHPPI outperforms state-of-the-art predictor by 6%, 2%, and 2% in terms of matthews correlation coefficient over 3 different benchmark viral-host PPI prediction datasets.

Objective Pakistan was one of the first South-Asian countries to introduce the ten-valent pneumococcal conjugate vaccine (PCV10) at the national level, using a 3+0 schedule without catchup, in 2013. Methods From 2014–18, fifteen children Results Pneumococcus was detected in 2370/3140 (75%) children. Vaccine type (VT) and non-vaccine type (NVT) serotypes were carried by 379 and 1990 children. There was a significant decline in VT carriage (by 40.3%, p-value Conclusion The emergence of NVT serotypes, primarily 19A replacing VT serotypes in this rural community, necessitates continuous monitoring of serotypes in the carriage and invasive disease to evaluate the utility of existing vaccine formulations.

Environmental enteric dysfunction (EED) is characterized by intestinal inflammation, malabsorption and growth-faltering in children with heightened exposure to gut pathogens. The aim of this study was to characterize serum non-esterified fatty acids (NEFA), in association with childhood undernutrition and EED, as potential biomarkers to predict growth outcomes. The study comprised a cohort of undernourished rural Pakistani infants (n = 365) and age-matched controls followed prospectively up to 24 months of age. Serum NEFA were quantified at ages 3–6 and 9 months and correlated with growth outcomes, serum bile acids and EED histopathological biomarkers. Serum NEFA correlated with linear growth-faltering and systemic and gut biomarkers of EED. Undernourished children exhibited essential fatty acid deficiency (EFAD), with low levels of linoleic acid and total n-6 polyunsaturated fatty acids, compensated by increased levels of oleic acid and increased elongase and desaturase activities. EFAD correlated with reduced anthropometric Z scores at 3–6 and 9 months of age. Serum NEFA also correlated with elevated BA and liver dysfunction. Essential fatty acid depletion and altered NEFA metabolism were highly prevalent and associated with acute and chronic growth-faltering in EED. The finding suggests that targeting early interventions to correct EFAD and promote FA absorption in children with EED may facilitate childhood growth in high-risk settings.

ABSTRACT Antimicrobial resistance is an emerging public health concern. Ten-valent pneumococcal vaccine (PCV10) was introduced in Pakistan’s Expanded Program on Immunization (EPI) in 2012 as a 3 + 0 schedule without catchup. From 2014 to 2018, children

This paper documents how an ethically aligned co-design methodology ensures trustworthiness in the early design phase of an artificial intelligence (AI) system component for healthcare. The system explains decisions made by deep learning networks analyzing images of skin lesions. The co-design of trustworthy AI developed here used a holistic approach rather than a static ethical checklist and required a multidisciplinary team of experts working with the AI designers and their managers. Ethical, legal, and technical issues potentially arising from the future use of the AI system were investigated. This paper is a first report on co-designing in the early design phase. Our results can also serve as guidance for other early-phase AI-similar tool developments.

Environmental enteric dysfunction (EED) is a subclinical condition of intestinal inflammation, barrier dysfunction and malabsorption associated with growth faltering in children living in poverty. This study explores association of altered duodenal permeability (lactulose, rhamnose and their ratio) with higher burden of enteropathogen in the duodenal aspirate, altered histopathological findings and higher morbidity (diarrhea) that is collectively associated with linear growth faltering in children living in EED endemic setting. In a longitudinal birth cohort, 51 controls (WHZ > 0, HAZ > -1.0) and 63 cases (WHZ< -2.0, refractory to nutritional intervention) were recruited. Anthropometry and morbidity were recorded on monthly bases up to 24 months of age. Dual sugar assay of urine collected after oral administration of lactulose and rhamnose was assessed in 96 children from both the groups. Duodenal histopathology (n = 63) and enteropathogen analysis of aspirate via Taqman array card (n = 60) was assessed in only cases. Giardia was the most frequent pathogen and was associated with raised L:R ratio (p = 0.068). Gastric microscopy was more sensitive than duodenal aspirate in H. pylori detection. Microscopically confirmed H. pylori negatively correlated with HAZ at 24 months (r = -0.313, p = 0.013). Regarding histopathological parameters, goblet cell reduction significantly correlated with decline in dual sugar excretion (p< 0.05). Between cases and controls, there were no significant differences in the median (25th, 75th percentile) of urinary concentrations (μg/ml) of lactulose [27.0 (11.50, 59.50) for cases vs. 38.0 (12.0, 61.0) for controls], rhamnose [66.0 (28.0, 178.0) vs. 86.5 (29.5, 190.5)] and L:R ratio [0.47 (0.24, 0.90) vs. 0.51 (0.31, 0.71)] respectively. In multivariable regression model, 31% of variability in HAZ at 24 months of age among cases and controls was explained by final model including dual sugars. In conclusion, enteropathogen burden is associated with altered histopathological features and intestinal permeability. In cases and controls living in settings of endemic enteropathy, intestinal permeability test may predict linear growth. However, for adoption as a screening tool for EED, further validation is required due to its complex intestinal pathophysiology.

Abstract Background Environmental Enteropathy (EE), characterized by alterations in intestinal structure, function, and immune activation, is believed to be an important contributor to childhood undernutrition and its associated morbidities, including stunting. Half of all global deaths in children 0) children. Blood, urine, and fecal samples, for evaluation of potential biomarkers, will be collected at various time points from all participants (longitudinal analyses). Participants will receive appropriate educational and nutritional interventions; non-responders will undergo further evaluation to determine eligibility for further workup, including upper gastrointestinal endoscopy. Histopathological changes in duodenal biopsies will be compared with duodenal biopsies obtained from USA controls who have celiac disease, Crohn’s disease, or who were found to have normal histopathology. RNA-Seq will be employed to characterize mucosal gene expression across groups. Duodenal biopsies, luminal aspirates from the duodenum, and fecal samples will be analyzed to define microbial community composition (omic analyses). The relationship between histopathology, mucosal gene expression, and community configuration will be assessed using a variety of bioinformatic tools to gain better understanding of disease pathogenesis and to identify mechanism-based biomarkers. Ethical review committees at all collaborating institutions have approved this study. All results will be made available to the scientific community. Discussion Operational and ethical constraints for safely obtaining intestinal biopsies from children in resource-poor settings have led to a paucity of human tissue-based investigations to understand and reverse EE in vulnerable populations. Furthermore, EE biomarkers have rarely been correlated with gold standard histopathological confirmation. The Study of Environmental Enteropathy and Malnutrition (SEEM) is designed to better understand the pathophysiology, predictors, biomarkers, and potential management strategies of EE to inform strategies to eradicate this debilitating pathology and accelerate progress towards the 2030 Sustainable Development Goals. Trial registration Retrospectively registered; clinicaltrials.gov ID NCT03588013.

Abstract Background With the advancement of powerful image processing and machine learning techniques, Computer Aided Diagnosis has become ever more prevalent in all fields of medicine including ophthalmology. These methods continue to provide reliable and standardized large scale screening of various image modalities to assist clinicians in identifying diseases. Since optic disc is the most important part of retinal fundus image for glaucoma detection, this paper proposes a two-stage framework that first detects and localizes optic disc and then classifies it into healthy or glaucomatous. Methods The first stage is based on Regions with Convolutional Neural Network (RCNN) and is responsible for localizing and extracting optic disc from a retinal fundus image while the second stage uses Deep Convolutional Neural Network to classify the extracted disc into healthy or glaucomatous. Unfortunately, none of the publicly available retinal fundus image datasets provides any bounding box ground truth required for disc localization. Therefore, in addition to the proposed solution, we also developed a rule-based semi-automatic ground truth generation method that provides necessary annotations for training RCNN based model for automated disc localization. Results The proposed method is evaluated on seven publicly available datasets for disc localization and on ORIGA dataset, which is the largest publicly available dataset with healthy and glaucoma labels, for glaucoma classification. The results of automatic localization mark new state-of-the-art on six datasets with accuracy reaching 100% on four of them. For glaucoma classification we achieved Area Under the Receiver Operating Characteristic Curve equal to 0.874 which is 2.7% relative improvement over the state-of-the-art results previously obtained for classification on ORIGA dataset. Conclusion Once trained on carefully annotated data, Deep Learning based methods for optic disc detection and localization are not only robust, accurate and fully automated but also eliminates the need for dataset-dependent heuristic algorithms. Our empirical evaluation of glaucoma classification on ORIGA reveals that reporting only Area Under the Curve, for datasets with class imbalance and without pre-defined train and test splits, does not portray true picture of the classifier’s performance and calls for additional performance metrics to substantiate the results.

Traditional distance and density-based anomaly detection techniques are unable to detect periodic and seasonality related point anomalies which occur commonly in streaming data, leaving a big gap in time series anomaly detection in the current era of the IoT. To address this problem, we present a novel deep learning-based anomaly detection approach (DeepAnT) for time series data, which is equally applicable to the non-streaming cases. DeepAnT is capable of detecting a wide range of anomalies, i.e., point anomalies, contextual anomalies, and discords in time series data. In contrast to the anomaly detection methods where anomalies are learned, DeepAnT uses unlabeled data to capture and learn the data distribution that is used to forecast the normal behavior of a time series. DeepAnT consists of two modules: time series predictor and anomaly detector. The time series predictor module uses deep convolutional neural network (CNN) to predict the next time stamp on the defined horizon. This module takes a window of time series (used as a context) and attempts to predict the next time stamp. The predicted value is then passed to the anomaly detector module, which is responsible for tagging the corresponding time stamp as normal or abnormal. DeepAnT can be trained even without removing the anomalies from the given data set. Generally, in deep learning-based approaches, a lot of data are required to train a model. Whereas in DeepAnT, a model can be trained on relatively small data set while achieving good generalization capabilities due to the effective parameter sharing of the CNN. As the anomaly detection in DeepAnT is unsupervised, it does not rely on anomaly labels at the time of model generation. Therefore, this approach can be directly applied to real-life scenarios where it is practically impossible to label a big stream of data coming from heterogeneous sensors comprising of both normal as well as anomalous points. We have performed a detailed evaluation of 15 algorithms on 10 anomaly detection benchmarks, which contain a total of 433 real and synthetic time series. Experiments show that DeepAnT outperforms the state-of-the-art anomaly detection methods in most of the cases, while performing on par with others.

Time series forecasting is one of the challenging problems for humankind. The traditional forecasting methods using mean regression models have severe shortcomings in reflecting real-world fluctuations. While new probabilistic methods rush to rescue, they fight with technical difficulties like quantile crossing or selecting a prior distribution. To meld the different strengths of these fields while avoiding their weaknesses, as well as, to push the boundary of the state-of-the-art, we introduce ForGAN - one step ahead probabilistic forecasting with generative adversarial networks. ForGAN utilizes the power of the conditional generative adversarial network to learn the data generating distribution and compute probabilistic forecasts from it. We argue how to evaluate ForGAN in opposition to regression methods. To investigate probabilistic forecasting of ForGAN, we create a new dataset and demonstrate our method abilities on it. This dataset will be made publicly available for comparison. Furthermore, we test ForGAN on two publicly available datasets, namely Mackey-Glass dataset and Internet traffic dataset (A5M), where the impressive performance of ForGAN demonstrate its high capability in forecasting future values.

This paper presents a novel framework for the demystification of convolutional deep learning models for time-series analysis. This is a step toward making informed/explainable decisions in the domain of time series, powered by deep learning. There have been numerous efforts to increase the interpretability of image-centric deep neural network models, where the learned features are more intuitive to visualize. Visualization in the domain of time series is significantly challenging, as there is no direct interpretation of the filters and inputs compared with the imaging modality. In addition, a little or no concentration has been devoted to the development of such tools in the domain of time series in the past. TSViz provides possibilities to explore and analyze the network from different dimensions at different levels of abstraction, which includes the identification of the parts of the input that were responsible for a particular prediction (including per filter saliency), importance of the different filters present in the network, notion of diversity present in the network through filter clustering, understanding of the main sources of variation learned by the network through inverse optimization, and analysis of the network's robustness against adversarial noise. As a sanity check for the computed influence values, we demonstrate our results on pruning of neural networks based on the computed influence information. These representations allow the user to better understand the network so that the acceptability of these deep models for time-series analysis can be enhanced. This is extremely important in domains, such as finance, industry 4.0, self-driving cars, health care, and counter-terrorism, where reasons for reaching a particular prediction are equally important as the prediction itself. We assess the proposed framework for interpretability with a set of desirable properties essential for any method in this direction.

The dataset described in this paper was collected for a time-series cross-sectional study exploring the impact of 10-valent Pneumococcal Conjugate Vaccine (PCV10) on nasopharyngeal (NP) carriage in children under 2 years of age from a rural population in Sindh, Pakistan. The study was carried out in two union councils of Matiari - Khyber and Shah Alam Shah Jee Wasi (Latitude 25.680298 / Longitude 68.502711). Data was collected on socio-demographics, clinical characteristics and vaccination status using android phone-based application. NP samples were collected using standard World Health Organisation (WHO) techniques, culture and serotyping was done using sequential Multiplex PCR described by Centre for Disease Control, USA. We looked at the carriage rate of vaccine type (VT) and non-vaccine type (NVT) serotypes over time in vaccinated and unvaccinated children. We additionally looked at the predictors for pneumococcal carriage. The uploaded dataset, available on Mendeley data repository (Nisar, Muhammad Imran (2021), “Impact of PCV10 on nasopharyngeal carriage in children in Pakistan”, Mendeley Data, V1, doi:10.17632/t79h6g97gr.1), has 3140 observations in CSV format. Additional files uploaded include a data dictionary and the set of questionnaires. The dataset and accompanying files can be used by other interested researchers to replicate our analysis, carry similar analysis under varying set of assumptions or perform additional exploratory or metanalysis

Pneumonia is the leading cause of morbidity and mortality in children worldwide. The ten valent pneumococcal vaccine (PCV10) was introduced in Pakistan's Expanded Program on Immunization (EPI) in 2012 as a 3 + 0 schedule without catchup immunization. Nasopharyngeal carriage is taken as a surrogate marker to measure the impact of pneumococcal vaccine on populations. Carriage surveys are necessary to monitor the persistence of Vaccine Type (VT) serotypes, the emergence of Non-Vaccine Type (NVT) serotypes, and their role in both transmission and disease. This article describes various troubleshooting measures which we undertook to adopt the protocol to our setting. We also used an innovative approach to describe various epidemiological parameters of vaccine effectiveness against carriage. It is important to publish these methods to allow for valid regional and temporal comparisons of results in different settings. Thus, in this article, we describe the following methods for isolating upper airway pneumococcal carriage: • Methods for the collection, transport, and storage of nasopharyngeal samples. • Methods for identification and serotyping of pneumococci. • Methods for estimation of the direct and indirect effects of pneumococcal vaccines on nasopharyngeal carriage.

The 10-valent pneumococcal vaccine was introduced in Pakistan’s Expanded Program on Immunization (EPI) in 2013 as a 3 + 0 schedule without catchup. We conducted three annual cross-sectional surveys from 2014–2016 to measure vaccine-type (VT) carriage in infants from a rural part of Pakistan. Nasopharyngeal specimens were collected by random sampling of infants from two union councils of Matiari. Samples were then transported to the Infectious Disease Research Laboratory (IDRL) at the Aga Khan University within 6–8 h of collection. Serotypes were established using sequential multiplex PCR. Of the 665 children enrolled across three surveys, 547 were culture-positive for pneumococcus. VT carriage decreased from 21.8% in 2014 to 12.7% in 2016 (p-value for trend

The recent developments in the machine-learning domain have enabled the development of complex multivariate probabilistic forecasting models. To evaluate the predictive power of these complex methods, it is pivotal to have a precise evaluation method to gauge the performance and predictability power of these complex methods. To do so, several evaluation metrics have been proposed in the past (such as the energy score, Dawid–Sebastiani score, and variogram score); however, these cannot reliably measure the performance of a probabilistic forecaster. Recently, CRPS-Sum has gained a lot of prominence as a reliable metric for multivariate probabilistic forecasting. This paper presents a systematic evaluation of CRPS-Sum to understand its discrimination ability. We show that the statistical properties of target data affect the discrimination ability of CRPS-Sum. Furthermore, we highlight that CRPS-Sum calculation overlooks the performance of the model on each dimension. These flaws can lead us to an incorrect assessment of model performance. Finally, with experiments on real-world datasets, we demonstrate that the shortcomings of CRPS-Sum provide a misleading indication of the probabilistic forecasting performance method. We illustrate that it is easily possible to have a better CRPS-Sum for a dummy model, which looks like random noise, in comparison to the state-of-the-art method.

Deep neural networks are one of the most successful classifiers across different domains. However, their use is limited in safety-critical areas due to their limitations concerning interpretability. The research field of explainable artificial intelligence addresses this problem. However, most interpretability methods align to the imaging modality by design. The paper introduces TimeREISE, a model agnostic attribution method that shows success in the context of time series classification. The method applies perturbations to the input and considers different attribution map characteristics such as the granularity and density of an attribution map. The approach demonstrates superior performance compared to existing methods concerning different well-established measurements. TimeREISE shows impressive results in the deletion and insertion test, Infidelity, and Sensitivity. Concerning the continuity of an explanation, it showed superior performance while preserving the correctness of the attribution map. Additional sanity checks prove the correctness of the approach and its dependency on the model parameters. TimeREISE scales well with an increasing number of channels and timesteps. TimeREISE applies to any time series classification network and does not rely on prior data knowledge. TimeREISE is suited for any usecase independent of dataset characteristics such as sequence length, channel number, and number of classes.

Studies conducted previously on the role-played by age in association with perceived organizational politics (POP) with different outcomes of job calls for more attention toward this domain. The current piece of research observes the effect of POP on job outcomes that include turnover intentions (intent to leave), job stress, and job satisfaction. It also sheds light on the age moderation in the POP association with job outcomes. The data used in this research were diverse and collected through self-administered and cross-sectional questionnaires. Multiple regression analysis was used as data analysis technique. Precisely, this research discovered that POP has a positive connection with job stress and intent to leave, while having no connection with job satisfaction. Moreover, POP impacts on job satisfaction and job stress was found to be positive in younger employees whereas in older employees, it affected both these variables negatively.

This study examines the adaptive market hypothesis (AMH) in relation to time-varying market efficiency by using three tests, namely Generalized Spectral (GS), Dominguez-Lobato (DL) and the automatic portmanteau test (AP) test on four-digital currencies; Bitcoin, Monaro, Litecoin, and Steller over the sample period of 2014–2018. The study applies Jarque-Bera test, ADF test, Ljung-Box statistics and ARCH-LM test for testing normality of returns, stationarity of series, serial correlation and volatility clustering in returns and squared returns of selected cryptocurrencies. Further, the study adopts an extremely important category of martingale difference hypothesis (MDH), which uses non-linear methods of dependencies for identifying changing linear and non-linear dependence in the price movement of currencies. The results indicate that price movements with linear and nonlinear dependences varies over time. Our tests also reveal that Bitcoin, Monaro and Litecoin have the longest efficiency periods. While Steller shows the longest inefficient market period. In view of varying market conditions, the results indicate that different market periods have significant impact on prices fluctuations of cryptocurrencies. Therefore, our findings suggest implementing the adaptive market hypothesis (AMH) as predicting changes in cryptocurrency prices over time must consider the time-varying market conditions for efficient forecasting.

We explore extreme return-volumes dependence among different cryptocurrencies such as Bitcoin, Ethereum, Ripple, and Litecoin by using the Copula approach. We use Student-t, Frank, Clayton, Survival Clayton, Gumbel, and SJC copulas. We filter out margins by using the EGARCH model for return series and GARCH model for volume series. Evidence of significant symmetric dependence between return-volume is not found due to insignificance of student-t and Frank copula parameters. In a return-volume relationship, coefficients of lower tail dependence are significant for Bitcoin, Ripple, and Litecoin which means that low returns are followed by low volumes. Lower tail dependence for the return-volume relationship is stronger than the upper tail dependence for Bitcoin, Ripple, and Litecoin. Moreover, for negative return-volume, left tail dependence coefficients are significant for Ripple and Litecoin, which means that high returns are followed by low volumes for Ripple and Litecoin. Our investigation shows that investors (buyer or seller) are very careful in extreme market conditions for both Ripple and Litecoin. Extreme upper tail and lower tail dependence coefficients are insignificant for Ethereum.