Progenitor cell isolation from mouse epididymal adipose tissue and sequencing library construction.

Here, we present a protocol to isolate progenitor cells from mouse epididymal visceral adipose tissue and construct bulk RNA and assay for transposase-accessible chromatin with sequencing (ATAC-seq) libraries. We describe steps for adipose tissue collection, cell isolation, and cell staining and sorting. We then detail procedures for both ATAC-seq and RNA sequencing library construction. This protocol can also be applied to other tissues and cell types directly or with minor modifications. For complete details on the use and execution of this protocol, please refer to Liu et al. (2023).1.

A single-cell atlas of bovine skeletal muscle reveals mechanisms regulating intramuscular adipogenesis and fibrogenesis.

BACKGROUND: Intramuscular fat (IMF) and intramuscular connective tissue (IMC) are often seen in human myopathies and are central to beef quality. The mechanisms regulating their accumulation remain poorly understood. Here, we explored the possibility of using beef cattle as a novel model for mechanistic studies of intramuscular adipogenesis and fibrogenesis. METHODS: Skeletal muscle single-cell RNAseq was performed on three cattle breeds, including Wagyu (high IMF), Brahman (abundant IMC but scarce IMF), and Wagyu/Brahman cross. Sophisticated bioinformatics analyses, including clustering analysis, gene set enrichment analyses, gene regulatory network construction, RNA velocity, pseudotime analysis, and cell-cell communication analysis, were performed to elucidate heterogeneities and differentiation processes of individual cell types and differences between cattle breeds. Experiments were conducted to validate the function and specificity of identified key regulatory and marker genes. Integrated analysis with multiple published human and non-human primate datasets was performed to identify common mechanisms. RESULTS: A total of 32 708 cells and 21 clusters were identified, including fibro/adipogenic progenitor (FAP) and other resident and infiltrating cell types. We identified an endomysial adipogenic FAP subpopulation enriched for COL4A1 and CFD (log2FC = 3.19 and 1.92, respectively; P < 0.0001) and a perimysial fibrogenic FAP subpopulation enriched for COL1A1 and POSTN (log2FC = 1.83 and 0.87, respectively; P < 0.0001), both of which were likely derived from an unspecified subpopulation. Further analysis revealed more progressed adipogenic programming of Wagyu FAPs and more advanced fibrogenic programming of Brahman FAPs. Mechanistically, NAB2 drives CFD expression, which in turn promotes adipogenesis. CFD expression in FAPs of young cattle before the onset of intramuscular adipogenesis was predictive of IMF contents in adulthood (R2  = 0.885, P < 0.01). Similar adipogenic and fibrogenic FAPs were identified in humans and monkeys. In aged humans with metabolic syndrome and progressed Duchenne muscular dystrophy (DMD) patients, increased CFD expression was observed (P < 0.05 and P < 0.0001, respectively), which was positively correlated with adipogenic marker expression, including ADIPOQ (R2  = 0.303, P < 0.01; and R2  = 0.348, P < 0.01, respectively). The specificity of Postn/POSTN as a fibrogenic FAP marker was validated using a lineage-tracing mouse line. POSTN expression was elevated in Brahman FAPs (P < 0.0001) and DMD patients (P < 0.01) but not in aged humans. Strong interactions between vascular cells and FAPs were also identified. CONCLUSIONS: Our study demonstrates the feasibility of beef cattle as a model for studying IMF and IMC. We illustrate the FAP programming during intramuscular adipogenesis and fibrogenesis and reveal the reliability of CFD as a predictor and biomarker of IMF accumulation in cattle and humans.

Intergenic transcription in in vivo developed bovine oocytes and pre-implantation embryos.

Background: Intergenic transcription, either failure to terminate at the transcription end site (TES), or transcription initiation at other intergenic regions, is present in cultured cells and enhanced in the presence of stressors such as viral infection. Transcription termination failure has not been characterized in natural biological samples such as pre-implantation embryos which express more than 10,000 genes and undergo drastic changes in DNA methylation. Results: Using Automatic Readthrough Transcription Detection (ARTDeco) and data of in vivo developed bovine oocytes and embryos, we found abundant intergenic transcripts that we termed as read-outs (transcribed from 5 to 15 kb after TES) and read-ins (transcribed 1 kb up-stream of reference genes, extending up to 15 kb up-stream). Read-throughs (continued transcription from TES of expressed reference genes, 4-15 kb in length), however, were much fewer. For example, the numbers of read-outs and read-ins ranged from 3,084 to 6,565 or 33.36-66.67% of expressed reference genes at different stages of embryo development. The less copious read-throughs were at an average of 10% and significantly correlated with reference gene expression (P < 0.05). Interestingly, intergenic transcription did not seem to be random because many intergenic transcripts (1,504 read-outs, 1,045 read-ins, and 1,021 read-throughs) were associated with common reference genes across all stages of pre-implantation development. Their expression also seemed to be regulated by developmental stages because many were differentially expressed (log2 fold change ≥ 2, P < 0.05). Additionally, while gradual but un-patterned decreases in DNA methylation densities 10 kb both up- and down-stream of the intergenic transcribed regions were observed, the correlation between intergenic transcription and DNA methylation was insignificant. Finally, transcription factor binding motifs and polyadenylation signals were found in 27.2% and 12.15% of intergenic transcripts, respectively, suggesting considerable novel transcription initiation and RNA processing. Conclusion: In summary, in vivo developed oocytes and pre-implantation embryos express large numbers of intergenic transcripts, which are not related to the overall DNA methylation profiles either up- or down-stream.

Tcf21 marks visceral adipose mesenchymal progenitors and functions as a rate-limiting factor during visceral adipose tissue development.

Distinct locations of different white adipose depots suggest anatomy-specific developmental regulation, a relatively understudied concept. Here, we report a population of Tcf21 lineage cells (Tcf21 LCs) present exclusively in visceral adipose tissue (VAT) that dynamically contributes to VAT development and expansion. During development, the Tcf21 lineage gives rise to adipocytes. In adult mice, Tcf21 LCs transform into a fibrotic or quiescent state. Multiomics analyses show consistent gene expression and chromatin accessibility changes in Tcf21 LC, based on which we constructed a gene-regulatory network governing Tcf21 LC activities. Furthermore, single-cell RNA sequencing (scRNA-seq) identifies the heterogeneity of Tcf21 LCs. Loss of Tcf21 promotes the adipogenesis and developmental progress of Tcf21 LCs, leading to improved metabolic health in the context of diet-induced obesity. Mechanistic studies show that the inhibitory effect of Tcf21 on adipogenesis is at least partially mediated via Dlk1 expression accentuation.

Completing Single-Cell DNA Methylome Profiles via Transfer Learning Together With KL-Divergence.

The high level of sparsity in methylome profiles obtained using whole-genome bisulfite sequencing in the case of low biological material amount limits its value in the study of systems in which large samples are difficult to assemble, such as mammalian preimplantation embryonic development. The recently developed computational methods for addressing the sparsity by imputing missing have their limits when the required minimum data coverage or profiles of the same tissue in other modalities are not available. In this study, we explored the use of transfer learning together with Kullback-Leibler (KL) divergence to train predictive models for completing methylome profiles with very low coverage (below 2%). Transfer learning was used to leverage less sparse profiles that are typically available for different tissues for the same species, while KL divergence was employed to maximize the usage of information carried in the input data. A deep neural network was adopted to extract both DNA sequence and local methylation patterns for imputation. Our study of training models for completing methylome profiles of bovine oocytes and early embryos demonstrates the effectiveness of transfer learning and KL divergence, with individual increase of 29.98 and 29.43%, respectively, in prediction performance and 38.70% increase when the two were used together. The drastically increased data coverage (43.80-73.6%) after imputation powers downstream analyses involving methylomes that cannot be effectively done using the very low coverage profiles (0.06-1.47%) before imputation.

Loss of Acta2 in cardiac fibroblasts does not prevent the myofibroblast differentiation or affect the cardiac repair after myocardial infarction.

In response to myocardial infarction (MI), quiescent cardiac fibroblasts differentiate into myofibroblasts mediating tissue repair. One of the most widely accepted markers of myofibroblast differentiation is the expression of Acta2 which encodes smooth muscle alpha-actin (SMαA) that is assembled into stress fibers. However, the requirement of Acta2/SMαA in the myofibroblast differentiation of cardiac fibroblasts and its role in post-MI cardiac repair remained unknown. To answer these questions, we generated a tamoxifen-inducible cardiac fibroblast-specific Acta2 knockout mouse line. Surprisingly, mice that lacked Acta2 in cardiac fibroblasts had a normal post-MI survival rate. Moreover, Acta2 deletion did not affect the function or histology of infarcted hearts. No difference was detected in the proliferation, migration, or contractility between WT and Acta2-null cardiac myofibroblasts. Acta2-null cardiac myofibroblasts had a normal total filamentous actin level and total actin level. Acta2 deletion caused a significant compensatory increase in the transcription level of non-Acta2 actin isoforms, especially Actg2 and Acta1. Moreover, in myofibroblasts, the transcription levels of cytoplasmic actin isoforms were significantly higher than those of muscle actin isoforms. In addition, we found that myocardin-related transcription factor-A is critical for myofibroblast differentiation but is not required for the compensatory effects of non-Acta2 isoforms. In conclusion, the Acta2 deletion does not prevent the myofibroblast differentiation of cardiac fibroblasts or affect the post-MI cardiac repair, and the increased expression and stress fiber formation of non-SMαA actin isoforms and the functional redundancy between actin isoforms are able to compensate for the loss of Acta2 in cardiac myofibroblasts.

Refinement of AlphaFold2 models against experimental and hybrid cryo-EM density maps.

Recent breakthroughs in deep learning-based protein structure prediction show that it is possible to obtain highly accurate models for a wide range of difficult protein targets for which only the amino acid sequence is known. The availability of accurately predicted models from sequences can potentially revolutionise many modelling approaches in structural biology, including the interpretation of cryo-EM density maps. Although atomic structures can be readily solved from cryo-EM maps of better than 4 Å resolution, it is still challenging to determine accurate models from lower-resolution density maps. Here, we report on the benefits of models predicted by AlphaFold2 (the best-performing structure prediction method at CASP14) on cryo-EM refinement using the Phenix refinement suite for AlphaFold2 models. To study the robustness of model refinement at a lower resolution of interest, we introduced hybrid maps (i.e. experimental cryo-EM maps) filtered to lower resolutions by real-space convolution. The AlphaFold2 models were refined to attain good accuracies above 0.8 TM scores for 9 of the 13 cryo-EM maps. TM scores improved for AlphaFold2 models refined against all 13 cryo-EM maps of better than 4.5 Å resolution, 8 hybrid maps of 6 Å resolution, and 3 hybrid maps of 8 Å resolution. The results show that it is possible (at least with the Phenix protocol) to extend the refinement success below 4.5 Å resolution. We even found isolated cases in which resolution lowering was slightly beneficial for refinement, suggesting that high-resolution cryo-EM maps might sometimes trap AlphaFold2 models in local optima.

The landscape of accessible chromatin in quiescent cardiac fibroblasts and cardiac fibroblasts activated after myocardial infarction.

After myocardial infarction, the massive death of cardiomyocytes leads to cardiac fibroblast proliferation and myofibroblast differentiation, which contributes to the extracellular matrix remodelling of the infarcted myocardium. We recently found that myofibroblasts further differentiate into matrifibrocytes, a newly identified cardiac fibroblast differentiation state. Cardiac fibroblasts of different states have distinct gene expression profiles closely related to their functions. However, the mechanism responsible for the gene expression changes during these activation and differentiation events is still not clear. In this study, the gene expression profiling and genome-wide accessible chromatin mapping of mouse cardiac fibroblasts isolated from the uninjured myocardium and the infarct at multiple time points corresponding to different differentiation states were performed by RNA sequencing (RNA-seq) and the assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-seq), respectively. ATAC-seq peaks were highly enriched in the promoter area and the distal area where the enhancers are located. A positive correlation was identified between the expression and promoter accessibility for many dynamically expressed genes, even though evidence showed that mechanisms independent of chromatin accessibility may also contribute to the gene expression changes in cardiac fibroblasts after MI. Moreover, motif enrichment analysis and gene regulatory network construction identified transcription factors that possibly contributed to the differential gene expression between cardiac fibroblasts of different states.

Highly Scalable Task Grouping for Deep Multi-Task Learning in Prediction of Epigenetic Events.

DNNs trained for predicting cellular events from DNA sequence have become emerging tools to help elucidate biological mechanisms underlying associations identified in genome-wide association studies. To enhance the training, multi-task learning (MTL) has been commonly exploited in previous works where trained networks were needed for multiple profiles differing in either event modality or cell type. All existing works adopted a simple MTL framework where all tasks share a single feature extraction network. Such a strategy even though effective to a certain extent leads to substantial negative transfer, meaning the existence of a large portion of tasks for which models obtained through MTL perform worse than those by single-task learning. There have been methods developed to address such negative transfer in other domains, such as computer vision. However, these methods are generally with limited scalability. In this paper, we propose a highly scalable task grouping framework to address negative transfer by only jointly training tasks that are potentially beneficial to each other. The proposed method exploits the network weights associated with task-specific classification heads that can be cheaply obtained by one-time joint training of all tasks. Our results using a dataset consisting of 367 epigenetic profiles demonstrate the effectiveness of the proposed approach and its superiority over baseline methods.

A Tool for Segmentation of Secondary Structures in 3D Cryo-EM Density Map Components Using Deep Convolutional Neural Networks.

Although cryo-electron microscopy (cryo-EM) has been successfully used to derive atomic structures for many proteins, it is still challenging to derive atomic structures when the resolution of cryo-EM density maps is in the medium resolution range, such as 5-10 Å. Detection of protein secondary structures, such as helices and ß-sheets, from cryo-EM density maps provides constraints for deriving atomic structures from such maps. As more deep learning methodologies are being developed for solving various molecular problems, effective tools are needed for users to access them. We have developed an effective software bundle, DeepSSETracer, for the detection of protein secondary structure from cryo-EM component maps in medium resolution. The bundle contains the network architecture and a U-Net model trained with a curriculum and gradient of episodic memory (GEM). The bundle integrates the deep neural network with the visualization capacity provided in ChimeraX. Using a Linux server that is remotely accessed by Windows users, it takes about 6 s on one CPU and one GPU for the trained deep neural network to detect secondary structures in a cryo-EM component map containing 446 amino acids. A test using 28 chain components of cryo-EM maps shows overall residue-level F1 scores of 0.72 and 0.65 to detect helices and ß-sheets, respectively. Although deep learning applications are built on software frameworks, such as PyTorch and Tensorflow, our pioneer work here shows that integration of deep learning applications with ChimeraX is a promising and effective approach. Our experiments show that the F1 score measured at the residue level is an effective evaluation of secondary structure detection for individual classes. The test using 28 cryo-EM component maps shows that DeepSSETracer detects ß-sheets more accurately than Emap2sec+, with a weighted average residue-level F1 score of 0.65 and 0.42, respectively. It also shows that Emap2sec+ detects helices more accurately than DeepSSETracer with a weighted average residue-level F1 score of 0.77 and 0.72 respectively.

The landscape of accessible chromatin in bovine oocytes and early embryos.

Chromatin reorganization governs the regulation of gene expression during preimplantation development. However, the landscape of chromatin dynamics in this period has not been explored in bovine. In this study, we constructed a genome-wide map of accessible chromatin in bovine oocytes and early embryos using an improved assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-seq) which revealed unique features of the accessible chromatin during bovine early embryo development. We found that chromatin accessibility is low in oocytes and 2-/4-cell embryos, followed by a significant increase in embryos during major embryonic genome activation (EGA), and peaked in elongating day 14 embryos. Genome-wide characteristics of open chromatin showed that ATAC-seq signals in both transcription start sites (TSS) and transcription end sites (TES) were strong. Additionally, the distal ATAC-seq peaks were enriched in repeat elements in a type-specific and stage-specific manner. We further unveiled a series of transcription factor (TF) motifs with distinct variation of enrichment from distal ATAC-seq peaks. By integrated analysis of chromatin accessibility with transcriptomes and DNA methylomes in bovine early embryos, we showed that promoter accessibility was positively correlated with gene expression, especially during major EGA, and was strongly correlated to DNA methylation and CpG density. Finally, we identified the critical chromatin signatures and TFs that differ between in vivo and in vitro derived blastocysts, which provides insights to the potential mechanisms leading to low quality of embryos produced in vitro. Together, this comprehensive analysis revealed critical features of chromatin landscape and epigenetic reprogramming during bovine preimplantation embryo development.

Transcription profiles of oocytes during maturation and embryos during preimplantation development in vivo in the goat.

RNA sequencing performed on goat matured oocytes and preimplantation embryos generated invivo enabled us to define the transcriptome for goat preimplantation embryo development. The largest proportion of changes in gene expression in goat was found at the 16-cell stage, not as previously defined at the 8-cell stage, and is later than in other mammalian species. In all, 6482 genes were identified to be significantly differentially expressed across all consecutive developmental stage comparisons, and the important signalling pathways involved in each development transition were determined. In addition, we identified genes that appear to be transcribed only at a specific stage of development. Using weighted gene coexpression network analysis, we found nine stage-specific modules of coexpressed genes that represent the corresponding stage of development. Furthermore, we identified conserved key members (or hub genes) of the goat transcriptional networks. Their association with other embryo genes suggests that they may have important regulatory roles in embryo development. Our cross-mammalian species transcriptomic comparisons demonstrate both conserved and goat-specific features of preimplantation development.

Using Curriculum Learning in Pattern Recognition of 3-dimensional Cryo-electron Microscopy Density Maps.

Although Cryo-electron microscopy (cryo-EM) has been successfully used to derive atomic structures for many proteins, it is still challenging to derive atomic structure when the resolution of cryo-EM density maps is in the medium range, e.g., 5-10 Å. Studies have attempted to utilize machine learning methods, especially deep neural networks to build predictive models for the detection of protein secondary structures from cryo-EM images, which ultimately helps to derive the atomic structure of proteins. However, the large variation in data quality makes it challenging to train a deep neural network with high prediction accuracy. Curriculum learning has been shown as an effective learning paradigm in machine learning. In this paper, we present a study using curriculum learning as a more effective way to utilize cryo-EM density maps with varying quality. We investigated three distinct training curricula that differ in whether/how images used for training in past are reused while the network was continually trained using new images. A total of 1,382 3-dimensional cryo-EM images were extracted from density maps of Electron Microscopy Data Bank in our study. Our results indicate learning with curriculum significantly improves the performance of the final trained network when the forgetting problem is properly addressed.

A genome-wide association study of cocaine use disorder accounting for phenotypic heterogeneity and gene­environment interaction

Background: Phenotypic heterogeneity and complicated gene­environment interplay in etiology are among the primary factors that hinder the identification of genetic variants associated with cocaine use disorder. Methods: To detect novel genetic variants associated with cocaine use disorder, we derived disease traits with reduced phenotypic heterogeneity using cluster analysis of a study sample (n = 9965). We then used these traits in genome-wide association tests, performed separately for 2070 African Americans and 1570 European Americans, using a new mixed model that accounted for the moderating effects of 5 childhood environmental factors. We used an independent sample (918 African Americans, 1382 European Americans) for replication. Results: The cluster analysis yielded 5 cocaine use disorder subtypes, of which subtypes 4 (n = 3258) and 5 (n = 1916) comprised heavy cocaine users, had high heritability estimates (h2 = 0.66 and 0.64, respectively) and were used in association tests. Seven of the 13 identified genetic loci in the discovery phase were available in the replication sample. In African Americans, rs114492924 (discovery p = 1.23 × E−8), a single nucleotide polymorphism in LINC01411, was replicated in the replication sample (p = 3.63 × E−3). In a meta-analysis that combined the discovery and replication results, 3 loci in African Americans were significant genome-wide: rs10188036 in TRAK2 (p = 2.95 × E−8), del-1:15511771 in TMEM51 (p = 9.11 × E−10) and rs149843442 near LPHN2 (p = 3.50 × E−8). Limitations: Lack of data prevented us from replicating 6 of the 13 identified loci. Conclusion: Our results demonstrate the importance of considering phenotypic heterogeneity and gene­environment interplay in detecting genetic variations that contribute to cocaine use disorder, because new genetic loci have been identified using our novel analytic method.

Gabapentin drug misuse signals: A pharmacovigilance assessment using the FDA adverse event reporting system.

BACKGROUND: Although there have been increasing reports of intentional gabapentin misuse, epidemiological evidence for the phenomenon is limited. The purpose of this study was to determine whether there are pharmacovigilance abuse signals for gabapentin. METHODS: Using FDA Adverse Events Reporting System reports from January 1, 2005 to December 31, 2015, we calculated pharmacovigilance signal measures (i.e., reporting odds ratio, proportional reporting ratio, information component, and empirical Bayes geometric mean) for abuse-related adverse event (AR-AE)-gabapentin pairs. Loglinear modeling assessed the frequency of concurrent reporting of abuse-related and abuse-specific AEs (AS-AEs) associated with gabapentin. Findings were compared to a positive (pregabalin) and negative (duloxetine) control. RESULTS: From 2005-2015 there were 5,951,229 unique AE reports submitted to the FDA including 99,977 for gabapentin, 73,977 for duloxetine, and 97,813 for pregabalin. Significant drug-AR-AE pair signals involving gabapentin included: drug abuser, multiple drug overdose, and substance-induced psychotic disorder. Significant drug AR-AE signals involving gabapentin and pregabalin, but not duloxetine, were: ataxia, dependence, drug abuse, increased drug tolerance, and overdose. Compared to duloxetine, gabapentin had significantly greater odds of a co-report for an AS-AE with drug withdrawal syndrome (OR: 6.55), auditory hallucinations (OR: 4.57), delusions (OR: 2.36), euphoric mood (OR: 5.45), ataxia (OR: 2.85), drug abuser (OR: 3.01), aggression (OR: 1.98), psychotic disorder (OR: 1.96), and feeling abnormal (OR: 1.31). CONCLUSIONS: We identified abuse-related signals for gabapentin and highlighted several CNS effects that may be associated with its abuse. Gabapentin prescribers should be aware of the drug's abuse liability and effects that may accompany its use.

Dynamics of trophoblast differentiation in peri-implantation-stage human embryos.

Single-cell RNA sequencing of cells from cultured human blastocysts has enabled us to define the transcriptomic landscape of placental trophoblast (TB) that surrounds the epiblast and associated embryonic tissues during the enigmatic day 8 (D8) to D12 peri-implantation period before the villous placenta forms. We analyzed the transcriptomes of 3 early placental cell types, cytoTB (CTB), syncytioTB (STB), and migratoryTB (MTB), picked manually from cultured embryos dissociated with trypsin and were able to follow sublineages that emerged from proliferating CTB at the periphery of the conceptus. A unique form of CTB with some features of STB was detectable at D8, while mature STB was at its zenith at D10. A form of MTB with a mixed MTB/CTB phenotype arose around D10. By D12, STB generation was in decline, CTB had entered a new phase of proliferation, and mature MTB cells had begun to move from the main body of the conceptus. Notably, the MTB transcriptome at D12 indicated enrichment of transcripts associated with IFN signaling, migration, and invasion and up-regulation of HLA-C, HLA-E, and HLA-G. The STB, which is distinct from the STB of later villous STB, had a phenotype consistent with intense protein export and placental hormone production, as well as migration and invasion. The studies show that TB associated with human embryos is in rapid developmental flux during peri-implantation period when it must invade, signal robustly to the mother to ensure that the pregnancy continues, and make first contact with the maternal immune system.

Multi-View Cluster Analysis with Incomplete Data to Understand Treatment Effects.

Multi-view cluster analysis, as a popular granular computing method, aims to partition sample subjects into consistent clusters across different views in which the subjects are characterized. Frequently, data entries can be missing from some of the views. The latest multi-view co-clustering methods cannot effectively deal with incomplete data, especially when there are mixed patterns of missing values. We propose an enhanced formulation for a family of multi-view co-clustering methods to cope with the missing data problem by introducing an indicator matrix whose elements indicate which data entries are observed and assessing cluster validity only on observed entries. In comparison with the simple strategy of removing subjects with missing values, our approach can use all available data in cluster analysis. In comparison with common methods that impute missing data in order to use regular multi-view analytics, our approach is less sensitive to imputation uncertainty. In comparison with other state-of-the-art multi-view incomplete clustering methods, our approach is sensible in the cases of missing any value in a view or missing the entire view, the most common scenario in practice. We first validated the proposed strategy in simulations, and then applied it to a treatment study of heroin dependence which would have been impossible with previous methods due to a number of missing-data patterns. Patients in a treatment study were naturally assessed in different feature spaces such as in the pre-, during-and post-treatment time windows. Our algorithm was able to identify subgroups where patients in each group showed similarities in all of the three time windows, thus leading to the recognition of pre-treatment (baseline) features predictive of post-treatment outcomes.

Inferring phenotypes from substance use via collaborative matrix completion.

BACKGROUND: Although substance use disorders (SUDs) are heritable, few genetic risk factors for them have been identified, in part due to the small sample sizes of study populations. To address this limitation, researchers have aggregated subjects from multiple existing genetic studies, but these subjects can have missing phenotypic information, including diagnostic criteria for certain substances that were not originally a focus of study. Recent advances in addiction neurobiology have shown that comorbid SUDs (e.g., the abuse of multiple substances) have similar genetic determinants, which makes it possible to infer missing SUD diagnostic criteria using criteria from another SUD and patient genotypes through statistical modeling. RESULTS: We propose a new approach based on matrix completion techniques to integrate features of comorbid health conditions and individual's genotypes to infer unreported diagnostic criteria for a disorder. This approach optimizes a bi-linear model that uses the interactions between known disease correlations and candidate genes to impute missing criteria. An efficient stochastic and parallel algorithm was developed to optimize the model with a speed 20 times greater than the classic sequential algorithm. It was tested on 3441 subjects who had both cocaine and opioid use disorders and successfully inferred missing diagnostic criteria with consistently better accuracy than other recent statistical methods. CONCLUSIONS: The proposed matrix completion imputation method is a promising tool to impute unreported or unobserved symptoms or criteria for disease diagnosis. Integrating data at multiple scales or from heterogeneous sources may help improve the accuracy of phenotype imputation.

[Clinical features of patients with Becker muscular dystrophy and deletions of the rod domain of dystrophin gene].

OBJECTIVE To explore the clinical features of patients carrying deletions of the rod domain of the dystrophin gene. METHODS Clinical data of 12 Chinese patients with Becker muscular dystrophy (BMD) and such deletions was reviewed. RESULTS Most patients complained of muscle weakness of lower limbs. Two patients had muscle cramps, one had increased creatine kinase (CK) level, and one had dilated cardiomyopathy. CONCLUSION Compared with DMD, the clinical features of BMD are much more variable, particularly for those carrying deletions of the rod domain of the dystrophin gene. Muscular weakness may not be the sole complaint of BMD. The diagnosis of BMD cannot be excluded by moderately elevated CK. For male patients with dilated cardiomyopathy, the possibility of BMD should be considered.

Collaborative Phenotype Inference from Comorbid Substance Use Disorders and Genotypes.

Data in large-scale genetic studies of complex human diseases, such as substance use disorders, are often incomplete. Despite great progress in genotype imputation, e.g., the IMPUTE2 method, considerably less progress has been made in inferring phenotypes. We designed a novel approach to integrate individuals' comorbid conditions with their genotype data to infer missing (unreported) diagnostic criteria of a disorder. The premise of our approach derives from correlations among symptoms and the shared biological bases of concurrent disorders such as co-dependence on cocaine and opioids. We describe a matrix completion method to construct a bi-linear model based on the interactions of genotypes and known symptoms of related disorders to infer unknown values of another set of symptoms or phenotypes. An efficient stochastic and parallel algorithm based on the linearized alternating direction method of multipliers was developed to solve the proposed optimization problem. Empirical evaluation of the approach in comparison with other advanced data matrix completion methods via a case study shows that it both significantly improves imputation accuracy and provides greater computational efficiency.