Search

Background and Aims The efficacy of statin therapy is hindered by intolerance to the therapy, leading to discontinuation. Variants in SLCO1B1, which encodes the hepatic transporter OATB1B1, influence statin pharmacokinetics, resulting in altered plasma concentration of the drug and its metabolites. Current pharmacogenetic guidelines require sequencing of the SLCO1B1 gene, which is more expensive and less accessible than genotyping. In this study we aimed to develop an easy, clinically implementable functional gene risk score (GRS) of common variants in SLCO1B1 to identify patients at risk of statin intolerance. Methods and Results A GRS was developed from four common variants in SLCO1B1. In statin users from Tayside Scotland, UK, those with a high-risk GRS had increased odds across three phenotypes of statin intolerance (general statin intolerance: ORGSI 2.42[95%CI:1.29, 4.31], p = 0.003; statin-related myopathy ORSRM 2.51[95%CI:1.28, 4.53], p = 0.004; statin-related suspected rhabdomyolysis: ORSRSR 2.85[95%CI:1.03, 6.65], p = 0.02). In contrast, using the Val174Ala genotype alone or the recommended OATP1B1 functional phenotypes produced weaker and less reliable results. A meta-analysis with results from adjudicated cases of statin-induced myopathy in the PREDICTION-ADR Consortium confirmed these findings (ORVal174Ala 1.99 [95%CI:1.01, 3.95], p = 0.048; ORGRS 1.76 [95%CI:1.16, 2.69], p = 0.008). For those requiring high-dose statin therapy, the high-risk GRS was more consistently associated with the time to onset of statin intolerance amongst the three phenotypes compared to Val174Ala (general statin intolerance: HRVal174Ala 2.49 [95%CI:1.09, 5.68], p = 0.03; HRGRS 2.44 [95%CI:1.46, 4.08], p Conclusions We provide evidence that a gene risk score based on four common SLCO1B1 variants provides an easily implemented genetic tool that is more reliable than the current recommended practice in estimating the risk and predicting early onset statin intolerance.

Aims/hypothesis South Asians in general, and Asian Indians in particular, have higher risk of type 2 diabetes compared with white Europeans, and a younger age of onset. The reasons for the younger age of onset in relation to obesity, beta cell function and insulin sensitivity are under-explored. Methods Two cohorts of Asian Indians, the ICMR-INDIAB cohort (Indian Council of Medical Research-India Diabetes Study) and the DMDSC cohort (Dr Mohan’s Diabetes Specialties Centre), and one of white Europeans, the ESDC (East Scotland Diabetes Cohort), were used. Using a cross-sectional design, we examined the comparative prevalence of healthy, overweight and obese participants with young-onset diabetes, classified according to their BMI. We explored the role of clinically measured beta cell function in diabetes onset in Asian Indians. Finally, the comparative distribution of a partitioned polygenic score (pPS) for risk of diabetes due to poor beta cell function was examined. Replication of the genetic findings was sought using data from the UK Biobank. Results The prevalence of young-onset diabetes with normal BMI was 9.3% amongst white Europeans and 24–39% amongst Asian Indians. In Asian Indians with young-onset diabetes, after adjustment for family history of type 2 diabetes, sex, insulin sensitivity and HDL-cholesterol, stimulated C-peptide was 492 pmol/ml (IQR 353–616, pp Conclusions/interpretation The prevalence of lean BMI in young-onset diabetes is over two times higher in Asian Indians compared with white Europeans. This phenotype of lean, young-onset diabetes appears driven in part by lower beta cell function. We demonstrate that Asian Indians with diabetes also have lower genetically determined beta cell function. Graphical abstract

Context A hypothesis-free genetic association analysis has not been reported for patients with primary hyperparathyroidism (PHPT). Objective We aimed to investigate genetic associations with PHPT using both genome-wide association study (GWAS) and candidate gene approaches. Methods A cross-sectional study was conducted among patients of European White ethnicity recruited in Tayside (Scotland, UK). Electronic medical records were used to identify PHPT cases and controls, and linked to genetic biobank data. Genetic associations were performed by logistic regression models and odds ratios (ORs). The combined effect of the genotypes was researched by genetic risk score (GRS) analysis. Results We identified 15 622 individuals for the GWAS that yielded 34 top single-nucleotide variations (formerly single-nucleotide polymorphisms), and LPAR3-rs147672681 reached genome-wide statistical significance (P = 1.2e-08). Using a more restricted PHPT definition, 8722 individuals with data on the GWAS-identified loci were found. Age- and sex-adjusted ORs for the effect alleles of SOX9-rs11656269, SLITRK5-rs185436526, and BCDIN3D-AS1-rs2045094 showed statistically significant increased risks (P < 1.5e-03). GRS analysis of 5482 individuals showed an OR of 2.51 (P = 1.6e-04), 3.78 (P = 4.0e-08), and 7.71 (P = 5.3e-17) for the second, third, and fourth quartiles, respectively, compared to the first, and there was a statistically significant linear trend across quartiles (P < 1.0e-04). Results were similar when stratifying by sex. Conclusion Using genetic loci discovered in a GWAS of PHPT carried out in a Scottish population, this study suggests new evidence for the involvement of genetic variants at SOX9, SLITRK5, LPAR3, and BCDIN3D-AS1. It also suggests that male and female carriers of greater numbers of PHPT-risk alleles both have a statistically significant increased risk of PHPT.

Generic pre-trained speech and text representations promise to reduce the need for large labeled datasets on specific speech and language tasks. However, it is not clear how to effectively adapt these representations for speech emotion recognition. Recent public benchmarks show the efficacy of several popular self-supervised speech representations for emotion classification. In this study, we show that the primary difference between the top-performing representations is in predicting valence while the differences in predicting activation and dominance dimensions are less pronounced. However, we show that even the best-performing HuBERT representation underperforms on valence prediction compared to a multimodal model that also incorporates text representation. We address this shortcoming by injecting lexical information into the speech representation using the multimodal model as a teacher. To improve the efficacy of our approach, we propose a novel estimate of the quality of the emotion predictions, to condition teacher-student training. We report new audio-only state-of-the-art concordance correlation coefficient (CCC) values of 0.757, 0.627, 0.671 for activation, valence and dominance predictions, respectively, on the MSP-Podcast corpus, and also state-of-the-art values of 0.667, 0.582, 0.545 on the IEMOCAP corpus., Accepted for publication at IEEE ICASSP 2022

Leukotrienes play a central pathophysiological role in both paediatric and adult asthma. However, 35% to 78% of asthmatics do not respond to leukotriene inhibitors. In this study we tested the role of theLTA4Hregulatory variant rs2660845 and age of asthma onset in response to montelukast in ethnically diverse populations. We identified and genotyped 3,594 asthma patients treated with montelukast (2,514 late-onset and 1,080 early-onset) from seven cohorts (UKBiobank, GoSHARE, BREATHE, Tayside RCT, PAGES, GALA II and SAGE). Individuals under montelukast treatment experiencing at least one exacerbation in a 12-month period were compared against individuals with no exacerbation, using logistic regression for each cohort and meta-analysis. While no significant association was found with European late-onset subjects, a meta-analysis of 523 early-onset individuals from European ancestry demonstrated the odds of experiencing asthma exacerbations by carriers of at least one G allele, despite montelukast treatment, were increased (odds-ratio = 2.92, 95%confidence interval (CI): 1.04–8.18, I2 = 62%, p = 0.0412) compared to those in the AA group. When meta-analysing with other ethnic groups, no significant increased risk of asthma exacerbations was found (OR = 1.60, 95% CI: 0.61–4.19, I2 = 85%, p = 0.342). Our study demonstrates that genetic variation inLTA4H, together with timing of asthma onset, may contribute to variability in montelukast response. European individuals with early-onset (≤18y) carrying at least one copy of rs2660845 have increased odd of exacerbation under montelukast treatment, presumably due to the up-regulation ofLTA4Hactivity. These findings support a precision medicine approach for the treatment of asthma with montelukast.

BackgroundSouth Asians have higher risk of type 2 diabetes compared to white Europeans and a younger age of onset. Reasons for the younger age of onset in relation to beta-cell function and insulin sensitivity are under-explored.MethodsTwo cohorts of Asian Indians, ICMR-INDIAB (Indian Council of Medical Research-INdia DIABetes Study) and DMDSC (Dr. Mohan’s Diabetes Specialties Centre) and one of white Europeans, ESDC (East Scotland Diabetes Cohort) were used. We examined the comparative prevalence of healthy, overweight, and obese BMI in young onset diabetes. We explored the role of clinically measured beta-cell function in diabetes onset in Asian Indians. Finally, the comparative distribution of a partitioned polygenic score (pPS) for risk of diabetes due to poor beta cell function was examined.ResultsPrevalence of young onset with normal BMI was 9.3% amongst white Europeans and 24%-39% amongst Asian Indians. In young diagnosed Asian Indians, after adjustment for family history of T2DM, sex, insulin sensitivity and HDL-c, stimulated C-peptide was 492pmol/mL (IQR: 353,616,PConclusionsAsian Indians have over two times the prevalence of lean BMI in young onset diabetes compared to white Europeans. This phenotype of lean, young onset diabetes appears driven in part by lower beta cell function. We demonstrate that Asian Indians with diabetes also have lower genetically determined beta cell function.

Speaker diarization accuracy can be affected by both acoustics and conversation characteristics. Determining the cause of diarization errors is difficult because speaker voice acoustics and conversation structure co-vary, and the interactions between acoustics, conversational structure, and diarization accuracy are complex. This paper proposes a methodology that can distinguish independent marginal effects of acoustic and conversation characteristics on diarization accuracy by remixing conversations in a factorial design. As an illustration, this approach is used to investigate gender-related and language-related accuracy differences with three diarization systems: a baseline system using subsegment x-vector clustering, a variant of it with shorter subsegments, and a third system based on a Bayesian hidden Markov model. Our analysis shows large accuracy disparities for the baseline system primarily due to conversational structure, which are partially mitigated in the other two systems. The illustration thus demonstrates how the methodology can be used to identify and guide diarization model improvements., 5 pages, 2 figures, Interspeech 2021

SUMMARYWe exploited genetic information to assess non-genetic influences in autoimmunity. We isolated gene modules whose products physically interact with environmental exposures related to autoimmunity, and analyzed their nominal statistical evidence of association with autoimmune and non-autoimmune diseases in genome-wide association studies (GWAS) data. Epstein Barr virus (EBV) and other Herpesviruses interactomes emerged as specifically associated with multiple sclerosis (MS), possibly under common regulatory mechanisms. Analyses of MS blood and brain transcriptomes, cytofluorimetric studies of endogenous EBV-infected lymphoblastoid lines, and lesion immunohistochemistry, confirmed a dysregulation of MS-associated EBV interactors, suggesting their contribution to CD40 signaling alterations in MS. These interactors resulted enriched in modules from inherited axonopathies-causing genes, supporting a link between EBV and neurodegeneration in MS, in accord with the observed transcriptomic dysregulations in MS brains. They were also enriched with top-ranked pharmaceutical targets prioritized on a genetic basis. This study delineates a disease-specific influence of herpesviruses on MS biology.

Training deep neural networks for automatic speech recognition (ASR) requires large amounts of transcribed speech. This becomes a bottleneck for training robust models for accented speech which typically contains high variability in pronunciation and other semantics, since obtaining large amounts of annotated accented data is both tedious and costly. Often, we only have access to large amounts of unannotated speech from different accents. In this work, we leverage this unannotated data to provide semantic regularization to an ASR model that has been trained only on one accent, to improve its performance for multiple accents. We propose Accent Pre-Training (Acc-PT), a semi-supervised training strategy that combines transfer learning and adversarial training. Our approach improves the performance of a state-of-the-art ASR model by 33% on average over the baseline across multiple accents, training only on annotated samples from one standard accent, and as little as 105 minutes of unannotated speech from a target accent.

Many of the recent advances in speech separation are primarily aimed at synthetic mixtures of short audio utterances with high degrees of overlap. Most of these approaches need an additional stitching step to stitch the separated speech chunks for long form audio. Since most of the approaches involve Permutation Invariant training (PIT), the order of separated speech chunks is nondeterministic and leads to difficulty in accurately stitching homogenous speaker chunks for downstream tasks like Automatic Speech Recognition (ASR). Also, most of these models are trained with synthetic mixtures and do not generalize to real conversational data. In this paper, we propose a speaker conditioned separator trained on speaker embeddings extracted directly from the mixed signal using an over-clustering based approach. This model naturally regulates the order of the separated chunks without the need for an additional stitching step. We also introduce a data sampling strategy with real and synthetic mixtures which generalizes well to real conversation speech. With this model and data sampling technique, we show significant improvements in speaker-attributed word error rate (SA-WER) on Hub5 data., Comment: Accepted for publication at Interspeech 2022

Leukotrienes play a central pathophysiological role in both pediatric and adult asthma. However, 35% to 78% of asthmatics do not respond to leukotriene inhibitors. To test the role of theLTA4Hregulatory variant rs2660845 and age of asthma onset in response to montelukast in ethnically diverse populations.We identified and genotyped 3,594 asthma patients treated with montelukast (2,514 late-onset and 1,080 early-onset) from seven cohorts (UKBiobank, GoSHARE, BREATHE, Tayside RCT, PAGES, GALA II and SAGE). Individuals under montelukast treatment experiencing at least one exacerbation in a 12-month period were compared against individuals with no exacerbation, using logistic regression for each cohort and meta-analysis.While no significant association was found with European late-onset subjects, a meta-analysis of 523 early-onset individuals from European ancestry demonstrated the risk of experiencing asthma exacerbations in the G allele carriers’ group (AG or GG), despite montelukast treatment, was increased (odds-ratio=3.27, 95%confidence interval: 0.98–10.93, I2=69%, p=0.05) compared to those in the AA group. When meta-analyzing with other ethnic groups, no significant increased risk of asthma exacerbations was found (OR=1.69, 95% CI: 0.56-5.09, I2=84.81%, p=0.35).Our study demonstrates that genetic variation inLTA4H, together with timing of asthma onset, may contribute to variability in montelukast response. Europeans individuals with early-onset (≤18y) carrying the rs2660845 G allele have increased risk of exacerbation under montelukast treatment, presumably due to the up-regulation ofLTA4Hactivity. These findings support a precision medicine approach for the treatment of asthma with montelukast.

SUMMARYWe exploited genetic information to assess the role of non-genetic factors in multifactorial diseases. To this aim we isolated candidate “interactomes” (i.e. groups of genes whose products are known to physically interact with environmental exposures and biological processes, plausibly relevant for disease pathogenesis) and analyzed nominal statistical evidence of association with genetic predisposition to multiple sclerosis (MS) and other inflammatory and non-inflammatory complex disorders. The interaction between genotype and Herpesviruses emerged as specific for MS, with Epstein Barr virus (EBV) showing higher levels of significance compared to Human Herpesvirus 8 (HHV8) and, more evidently, to cytomegalovirus (CMV). In accord with this result, when we classified the MS-associated genes contained in the interactomes into canonical pathways, the analysis converged towards biological functions of B cells, in particular the CD40 pathway. When we analyzed peripheral blood transcriptomes in persons with MS, we found a significant dysregulation of MS-associated genes belonging to the EBV interactome in primary progressive MS. This study indicates that the interaction between herpesviruses and predisposing genetic background is of causal significance in MS, and provides a mechanistic explanation for the long-recognized association between EBV and this condition.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

In multiple sclerosis, the pathological interaction between autoreactive Th cells and mononuclear phagocytes in the CNS drives initiation and maintenance of chronic neuroinflammation. Here, we found that intrathecal transplantation of neural stem/precursor cells (NPCs) in mice with experimental autoimmune encephalomyelitis (EAE) impairs the accumulation of inflammatory monocyte-derived cells (MCs) in the CNS, leading to improved clinical outcome. Secretion of IL-23, IL-1, and TNF-α, the cytokines required for terminal differentiation of Th cells, decreased in the CNS of NPC-treated mice, consequently inhibiting the induction of GM-CSF-producing pathogenic Th cells. In vivo and in vitro transcriptome analyses showed that NPC-secreted factors inhibit MC differentiation and activation, favoring the switch toward an antiinflammatory phenotype. Tgfb2-/-NPCs transplanted into EAE mice were ineffective in impairing MC accumulation within the CNS and failed to drive clinical improvement. Moreover, intrathecal delivery of TGF-β2 during the effector phase of EAE ameliorated disease severity. Taken together, these observations identify TGF-β2 as the crucial mediator of NPC immunomodulation. This study provides evidence that intrathecally transplanted NPCs interfere with the CNS-restricted inflammation of EAE by reprogramming infiltrating MCs into antiinflammatory myeloid cells via secretion of TGF-β2.

Hidden Markov models (HMMs) with Gaussian mixture distributions rely on an assumption that speech features are temporally uncorrelated, and often assume a diagonal covariance matrix where correlations between feature vectors for adjacent frames are ignored. A Linear Dynamic Model (LDM) is a Markovian state-space model that also relies on hidden state modeling, but explicitly models the evolution of these hidden states using an autoregressive process. An LDM is capable of modeling higher order statistics and can exploit correlations of features in an efficient and parsimonious manner. In this paper, we present a hybrid LDM/HMM decoder architecture that postprocesses segmentations derived from the first pass of an HMM-based recognition. This smoothed trajectory model is complementary to existing HMM systems. An Expectation-Maximization (EM) approach for parameter estimation is presented. We demonstrate a 13 % relative WER reduction on the Aurora-4 clean evaluation set, and a 13 % relative WER reduction on the babble noise condition.

Intracranial large vessel involvement is an unusual complication of tuberculous meningitis. The authors report a 39-year-old female presenting with an episode of seizure, followed by rapid decline in sensorium without prominent systemic features. An initial cranial magnetic resonance imaging revealed tuberculomata and patchy infarcts. Despite antituberculous therapy, she progressively worsened. A cranial computed tomography scan done following the worsening revealed a massive middle-cerebral artery (MCA) infarct. Unfortunately, the patient died in spite of decompressive craniotomy. Malignant MCA territory infarct is a rare and potentially fatal complication of tuberculous meningitis.