The single-cell opioid responses in the context of HIV (SCORCH) consortium.

Substance use disorders (SUD) and drug addiction are major threats to public health, impacting not only the millions of individuals struggling with SUD, but also surrounding families and communities. One of the seminal challenges in treating and studying addiction in human populations is the high prevalence of co-morbid conditions, including an increased risk of contracting a human immunodeficiency virus (HIV) infection. Of the ~15 million people who inject drugs globally, 17% are persons with HIV. Conversely, HIV is a risk factor for SUD because chronic pain syndromes, often encountered in persons with HIV, can lead to an increased use of opioid pain medications that in turn can increase the risk for opioid addiction. We hypothesize that SUD and HIV exert shared effects on brain cell types, including adaptations related to neuroplasticity, neurodegeneration, and neuroinflammation. Basic research is needed to refine our understanding of these affected cell types and adaptations. Studying the effects of SUD in the context of HIV at the single-cell level represents a compelling strategy to understand the reciprocal interactions among both conditions, made feasible by the availability of large, extensively-phenotyped human brain tissue collections that have been amassed by the Neuro-HIV research community. In addition, sophisticated animal models that have been developed for both conditions provide a means to precisely evaluate specific exposures and stages of disease. We propose that single-cell genomics is a uniquely powerful technology to characterize the effects of SUD and HIV in the brain, integrating data from human cohorts and animal models. We have formed the Single-Cell Opioid Responses in the Context of HIV (SCORCH) consortium to carry out this strategy.

iHerd: an integrative hierarchical graph representation learning framework to quantify network changes and prioritize risk genes in disease.

Different genes form complex networks within cells to carry out critical cellular functions, while network alterations in this process can potentially introduce downstream transcriptome perturbations and phenotypic variations. Therefore, developing efficient and interpretable methods to quantify network changes and pinpoint driver genes across conditions is crucial. We propose a hierarchical graph representation learning method, called iHerd. Given a set of networks, iHerd first hierarchically generates a series of coarsened sub-graphs in a data-driven manner, representing network modules at different resolutions (e.g., the level of signaling pathways). Then, it sequentially learns low-dimensional node representations at all hierarchical levels via efficient graph embedding. Lastly, iHerd projects separate gene embeddings onto the same latent space in its graph alignment module to calculate a rewiring index for driver gene prioritization. To demonstrate its effectiveness, we applied iHerd on a tumor-to-normal GRN rewiring analysis and cell-type-specific GCN analysis using single-cell multiome data of the brain. We showed that iHerd can effectively pinpoint novel and well-known risk genes in different diseases. Distinct from existing models, iHerd's graph coarsening for hierarchical learning allows us to successfully classify network driver genes into early and late divergent genes (EDGs and LDGs), emphasizing genes with extensive network changes across and within signaling pathway levels. This unique approach for driver gene classification can provide us with deeper molecular insights. The code is freely available at https://github.com/aicb-ZhangLabs/iHerd. All other relevant data are within the manuscript and supporting information files.

Resolving misalignment interference for NGS-based clinical diagnostics.

Next-generation sequencing (NGS) is an incredibly useful tool for genetic disease diagnosis. However, the most commonly used bioinformatics methods for analyzing sequence reads insufficiently discriminate genomic regions with extensive sequence identity, such as gene families and pseudogenes, complicating diagnostics. This problem has been recognized for specific genes, including many involved in human disease, and diagnostic labs must perform additional costly steps to guarantee accurate diagnosis in these cases. Here we report a new data analysis method based on the comparison of read depth between highly homologous regions to identify misalignment. Analyzing six clinically important genes-CYP21A2, GBA, HBA1/2, PMS2, and SMN1-each exhibiting misalignment issues related to homology, we show that our technique can correctly identify potential misalignment events and be used to make appropriate calls. Combined with long-range PCR and/or MLPA orthogonal testing, our clinical laboratory can improve variant calling with minimal additional cost. We propose an accurate and cost-efficient NGS testing procedure that will benefit disease diagnostics, carrier screening, and research-based population studies.

Alternative polyadenylation sites reveal distinct chromatin accessibility and histone modification in human cell lines.

MOTIVATION: In addition to alternative splicing, alternative polyadenylation has also been identified as a critical and prevalent regulatory mechanism in human gene expression. However, the mechanism of alternative polyadenylation selection and the involved factors is still largely unknown. RESULTS: We use the ENCODE data to scan DNA functional elements, including chromatin accessibility and histone modification, around transcript cleavage sites. Our results demonstrate that polyadenylation sites tend to be less sensitive to DNase I. However, these polyadenylation sites have preference in nucleosome-depleted regions, indicating the involvement of chromatin higher-order structure rather than nucleosomes in the resultant lower chromatin accessibility. More interestingly, for genes using two polyadenylation sites, the distal sites show even lower chromatin accessibility compared with the proximal sites or the unique sites of genes using only one polyadenylation site. We also observe that the histone modification mark, histone H3 lysine 36 tri-methylation (H3K36Me3), exhibits different patterns around the cleavage sites of genes using multiple polyadenylation sites from those of genes using a single polyadenylation site. Surprisingly, the H3K36Me3 levels are comparable among the alternative polyadenylation sites themselves. In summary, polyadenylation and alternative polyadenylation are closely related to functional elements on the DNA level. CONTACT: liang.chen@usc.edu.

Turtling: a time-aware neural topic model on NIH grant data.

Motivation: Recent initiatives for federal grant transparency allow direct knowledge extraction from large volumes of grant texts, serving as a powerful alternative to traditional surveys. However, its computational modeling is challenging as grants are usually multifaceted with constantly evolving topics. Results: We propose Turtling, a time-aware neural topic model with three unique characteristics. First, Turtling employs pretrained biomedical word embedding to extract research topics. Second, it leverages a probabilistic time-series model to allow smooth and coherent topic evolution. Lastly, Turtling leverages additional topic diversity loss and funding institute classification loss to improve topic quality and facilitate funding institute prediction. We apply Turtling on publicly available NIH grant text and show that it significantly outperforms other methods on topic quality metrics. We also demonstrate that Turtling can provide insights into research topic evolution by detecting topic trends across decades. In summary, Turtling may be a valuable tool for grant text analysis. Availability and implementation: Turtling is freely available as an open-source software at https://github.com/aicb-ZhangLabs/Turtling.

Beneficial Effects of Oral Nutrition Supplements on the Nutritional Status and Physical Performance of Older Nursing Home Residents at Risk of Malnutrition.

The purpose of this study was to compare the effects of nutritional supplement drinks (NSDs) and nutritional education (NE) on the nutritional status and physical performance of older nursing home residents who were at risk of malnutrition. This study was a clustered, randomized, parallel, multi-center clinical trial, with 107 participants more than 65 years old and at risk of malnutrition recruited from several nursing homes in this study. Participants were divided into two groups: an NE group (n = 50) and an NSD group (n = 57). The NE group was given NE by a dietitian, while the NSD group was provided with two packs of NSD except receiving NE (Mei Balance, Meiji Holdings, Tokyo, Japan) per day as a snack between meals and before bed. Anthropometric data, blood pressure, nutritional status, blood biochemical biomarkers, and physical performance were measured before and after 12-week interventions. After 12 weeks of NE combined with NSD intervention, body weight, body-mass index, the mini nutritional assessment-short form (MNA-SF) score, walking speed, and SF-36 questionnaire score were improved in older nursing home residents at risk of malnutrition.

First Trimester Prediction of Preterm Birth in Patient Plasma with Machine-Learning-Guided Raman Spectroscopy and Metabolomics.

Preterm birth (PTB) is the leading cause of infant deaths globally. Current clinical measures often fail to identify women who may deliver preterm. Therefore, accurate screening tools are imperative for early prediction of PTB. Here, we show that Raman spectroscopy is a promising tool for studying biological interfaces, and we examine differences in the maternal metabolome of the first trimester plasma of PTB patients and those that delivered at term (healthy). We identified fifteen statistically significant metabolites that are predictive of the onset of PTB. Mass spectrometry metabolomics validates the Raman findings identifying key metabolic pathways that are enriched in PTB. We also show that patient clinical information alone and protein quantification of standard inflammatory cytokines both fail to identify PTB patients. We show for the first time that synergistic integration of Raman and clinical data guided with machine learning results in an unprecedented 85.1% accuracy of risk stratification of PTB in the first trimester that is currently not possible clinically. Correlations between metabolites and clinical features highlight the body mass index and maternal age as contributors of metabolic rewiring. Our findings show that Raman spectral screening may complement current prenatal care for early prediction of PTB, and our approach can be translated to other patient-specific biological interfaces.

Translator: A Transfer Learning Approach to Facilitate Single-Cell ATAC-Seq Data Analysis from Reference Dataset.

Recent advances in single-cell sequencing assay for transposase-accessible chromatin (scATAC-seq) have allowed simultaneous epigenetic profiling over thousands of individual cells to dissect the cellular heterogeneity and elucidate regulatory mechanisms at the finest possible resolution. However, scATAC-seq is challenging to model computationally due to the ultra-high dimensionality, low signal-to-noise ratio, complex feature interactions, and high vulnerability to various confounding factors. In this study, we present Translator, an efficient transfer learning approach to capture generalizable chromatin interactions from high-quality (HQ) reference scATAC-seq data to obtain robust cell representations in low-to-moderate quality target scATAC-seq data. We applied Translator on various simulated and real scATAC-seq datasets and demonstrated that Translator could learn more biologically meaningful cell representations than other methods by incorporating information learned from the reference data, thus facilitating various downstream analyses such as clustering and motif enrichment measurements. Moreover, Translator's block-wise deep learning framework can handle nonlinear relationships with restricted connections using fewer parameters to boost computational efficiency through Graphics Processing Unit (GPU) parallelism. Finally, we have implemented Translator as a free software package available for the community to leverage large-scale, HQ reference data to study target scATAC-seq data.

CPDB: a database of circular permutation in proteins.

Circular permutation (CP) in a protein can be considered as if its sequence were circularized followed by a creation of termini at a new location. Since the first observation of CP in 1979, a substantial number of studies have concluded that circular permutants (CPs) usually retain native structures and functions, sometimes with increased stability or functional diversity. Although this interesting property has made CP useful in many protein engineering and folding researches, large-scale collections of CP-related information were not available until this study. Here we describe CPDB, the first CP DataBase. The organizational principle of CPDB is a hierarchical categorization in which pairs of circular permutants are grouped into CP clusters, which are further grouped into folds and in turn classes. Additions to CPDB include a useful set of tools and resources for the identification, characterization, comparison and visualization of CP. Besides, several viable CP site prediction methods are implemented and assessed in CPDB. This database can be useful in protein folding and evolution studies, the discovery of novel protein structural and functional relationships, and facilitating the production of new CPs with unique biotechnical or industrial interests. The CPDB database can be accessed at http://sarst.life.nthu.edu.tw/cpdb.

iSARST: an integrated SARST web server for rapid protein structural similarity searches.

iSARST is a web server for efficient protein structural similarity searches. It is a multi-processor, batch-processing and integrated implementation of several structural comparison tools and two database searching methods: SARST for common structural homologs and CPSARST for homologs with circular permutations. iSARST allows users submitting multiple PDB/SCOP entry IDs or an archive file containing many structures. After scanning the target database using SARST/CPSARST, the ordering of hits are refined with conventional structure alignment tools such as FAST, TM-align and SAMO, which are run in a PC cluster. In this way, iSARST achieves a high running speed while preserving the high precision of refinement engines. The final outputs include tables listing co-linear or circularly permuted homologs of the query proteins and a functional summary of the best hits. Superimposed structures can be examined through an interactive and informative visualization tool. iSARST provides the first batch mode structural comparison web service for both co-linear homologs and circular permutants. It can serve as a rapid annotation system for functionally unknown or hypothetical proteins, which are increasing rapidly in this post-genomics era. The server can be accessed at http://sarst.life.nthu.edu.tw/iSARST/.

The Value of Parental Testing by Next-Generation Sequencing Includes the Detection of Germline Mosaicism.

When a potential disease-causing variant is detected in a proband, parental testing is used to determine the mode of inheritance. This study demonstrates that next-generation sequencing (NGS) is uniquely well suited for parental testing, in particular because of its ability to detect clinically relevant germline mosaicism. Parental variant testing by NGS was performed in a clinical laboratory for 1 year. The detection of mosaicism by NGS was compared with its detection by Sanger sequencing. Eight cases of previously unrevealed mosaicism were detected by NGS across eight different genes. Mosaic variants were differentiated from sequencing noise using custom bioinformatics analyses in combination with familial inheritance data and complementary Sanger sequencing. Sanger sequencing detected mosaic variants with allele fractions ≥8% by NGS, but could not detect mosaic variants below that level. Detection of germline mosaicism by NGS is invaluable to parents, providing a more accurate recurrence risk that can alter decisions on family planning and pregnancy management. Because NGS can also confirm parentage and increase scalability, it simultaneously streamlines and strengthens the variant curation process. These features make NGS the ideal method for parental testing, superior even to Sanger sequencing for most genomic loci.