Biology subjects

Baras, A.

Publications and source records attributed to Baras, A..

16 recordsLinked to original sources

bioRxiv2025

Applying Self-Supervised Tissue Patch Encoders to Enable Cellular Graph Encoding

Recent advances in computation pathology have seen the development of various forms of foundational models that have enabled high-quality, generalpurpose feature extraction from tissue patches. However, most of these models are somewhat limited in their ability to capture cell-to-cell spatial relationships essential for understanding tissue microenvironments. To address this issue, self-supervised learning (SSL) applied to cellular graph models is a promising avenue of investigation as this would afford one the ability to capture cell-level interpretability of the tissue microenvironment and learn relationships from the cellular topology. Compared to SSL on tissue patches, developing cellular graph models requires considerable computation and algorithmic effort, starting with accurate cell detection/segmentation from tissue and then subsequently requiring adequate cell-level feature extraction. In this study, we demonstrate that a pretrained tissue patch encoder, originally trained in the context of the token used in conventional vision transformer frameworks can also serve effectively as a cellular graph encoder for cellbased patches. We report three key findings: 1) the pretrained tissue patch encoder acts as a powerful graph-level feature extractor for cellular graphs in tissue patches; 2) it also can function as a celllevel feature extractor, allowing each cell to be contextualized within its local tissue microenvironment environment; 3) the cell patch size and positional embeddings are critical factors for high-quality cellular graph feature extraction. To validate and support these insights, we evaluate our approach on various tissue region classification tasks from three different types of cancers. Experimental results demonstrate that our method achieves performance comparable to the current state-of-the-art foundation models in computational pathology. These results highlight the potential of pretrained tissue patch encoders in cellular graph modeling, which can enable insights into how the cellular features and topology are related to various biological processes and phenomena.

bioinformatics↗

bioRxiv2025

Application and Characterization of the Multiple Instance Learning Framework in Flow Cytometry

For decades, flow cytometry has allowed for single-cell profiling based on selected biomarkers and is widely used in both clinical and research settings. One major limitation of most conventional flow cytometry analyses is the dependency on a mostly manual gating process. This generally involves sequentially selecting biomarkers to isolate phenotype-associated cell populations, an approach that is both labor-intensive and prone to bias. To address this challenge, we introduce the application of a series of multi-instance learning frameworks for automated flow cytometry data analysis. Our models demonstrate strong performance across diverse biomedical applications, including cancer subtyping based on tumor-infiltrating immune cells, HIV survival stratification, AML minimal residual disease prediction, and COVID-19 severity assessment. We further examine how network architecture affects predictive performance and the detection of rare but clinically significant cell populations. Notably, our models utilize attention mechanisms to directly identify phenotype-associated cell subsets, serving as an interpretable, data-driven alternative to fully manual gating. These findings underscore the potential of multi-instance learning as a scalable and interpretable framework for flow cytometry, with broad applications in precision medicine and translational immunology.

bioinformatics↗

bioRxiv2025

Variant Classification Using Proteomics-Informed Large Language Models Increases Power of Rare Variant Association Studies and Enhances Target Discovery

Rare variant association analysis, which assesses the aggregate effect of rare damaging variants within a gene, is a powerful strategy for advancing knowledge of human biology. Numerous models have been proposed to identify damaging coding variants, with the most recent ones employing deep learning and large language models (LLM) to predict the impact of changes in coding sequences. Here, we use newly available proteomics data on 2,898 proteins across 46, 665 individuals to evaluate and refine LLM predictors of damaging variants. Using one of these refined models, we evaluate association between rare damaging variants and human phenotypes at 241 positive control gene-trait pairs. Among these gene-trait pairs, our proteomics-guided model outperforms an ensemble of conventional approaches including PolyPhen2, Mutation Taster, SIFT, and LRT, as well as newer machine learning approaches for identifying damaging missense variants, such as CADD, ESM-1v, ESM-1b and AlphaMissense. When attempting to recover known associations by correctly separating damaging singleton missense variants from other singleton variants, our approach recapitulates 36.5% of gene-trait pairs with known associations, exceeding all the alternatives we considered. Furthermore, when we apply our model to 10 exemplary traits from the UK Biobank, we identify 177 gene-trait associations - again exceeding all other approaches. Our results demonstrate that summary statistics from large-scale human proteomics data enable evaluation and refinement of coding variant classification LLMs, improving discovery potential in human genetic studies.

bioinformatics↗

bioRxiv2023

A deep catalog of protein-coding variation in 985,830 individuals

Coding variants that have significant impact on function can provide insights into the biology of a gene but are typically rare in the population. Identifying and ascertaining the frequency of such rare variants requires very large sample sizes. Here, we present the largest catalog of human protein-coding variation to date, derived from exome sequencing of 985,830 individuals of diverse ancestry to serve as a rich resource for studying rare coding variants. Individuals of African, Admixed American, East Asian, Middle Eastern, and South Asian ancestry account for 20% of this Exome dataset. Our catalog of variants includes approximately 10.5 million missense (54% novel) and 1.1 million predicted loss-of-function (pLOF) variants (65% novel, 53% observed only once). We identified individuals with rare homozygous pLOF variants in 4,874 genes, and for 1,838 of these this work is the first to document at least one pLOF homozygote. Additional insights from the RGC-ME dataset include 1) improved estimates of selection against heterozygous loss-of-function and identification of 3,459 genes intolerant to loss-of-function, 83 of which were previously assessed as tolerant to loss-of-function and 1,241 that lack disease annotations; 2) identification of regions depleted of missense variation in 457 genes that are tolerant to loss-of-function; 3) functional interpretation for 10,708 variants of unknown or conflicting significance reported in ClinVar as cryptic splice sites using splicing score thresholds based on empirical variant deleteriousness scores derived from RGC-ME; and 4) an observation that approximately 3% of sequenced individuals carry a clinically actionable genetic variant in the ACMG SF 3.1 list of genes. We make this important resource of coding variation available to the public through a variant allele frequency browser. We anticipate that this report and the RGC-ME dataset will serve as a valuable reference for understanding rare coding variation and help advance precision medicine efforts.

genetics↗

bioRxiv2023

Joint testing of rare variant burden scores using non-negative least squares

Gene-based burden tests are a popular and powerful approach for analysis of exome-wide association studies. These approaches combine sets of variants within a gene into a single burden score that is then tested for association. Typically, a range of burden scores are calculated and tested across a range of annotation classes and frequency bins. Correlation between these tests can complicate the multiple testing correction and hamper interpretation of the results. We introduce a new method called the Sparse Burden Association Test (SBAT) that tests the joint set of burden scores under the assumption that causal burden scores act in the same effect direction. The method simultaneously assesses the significance of the model fit and selects the set of burden scores that best explain the association at the same time. Using simulated data, we show that the method is well calibrated and highlight some scenarios where the test outperforms existing gene-based tests. We apply the method to 73 quantitative traits from the UK Biobank which further illustrates the power of the method. This test is implemented in the REGENIE software.

genetics↗

bioRxiv2022

Genotyping, sequencing and analysis of 140,000 adults from the Mexico City Prospective Study

The Mexico City Prospective Study (MCPS) is a prospective cohort of over 150,000 adults recruited two decades ago from the urban districts of Coyoacan and Iztapalapa in Mexico City. We generated genotype and exome sequencing data for all individuals, and whole genome sequencing for 10,000 selected individuals. We uncovered high levels of relatedness and substantial heterogeneity in ancestry composition across individuals. Most sequenced individuals had admixed Native American, European and African ancestry, with extensive admixture from indigenous groups in Central, Southern and South Eastern Mexico. Native Mexican segments of the genome had lower levels of coding variation, but an excess of homozygous loss of function variants compared with segments of African and European origin. We estimated population specific allele frequencies at 142 million genomic variants, with an effective sample size of 91,856 for Native Mexico at exome variants, all available via a public browser. Using whole genome sequencing, we developed an imputation reference panel which outperforms existing panels at common variants in individuals with high proportions of Central, South and South Eastern Native Mexican ancestry. Our work illustrates the value of genetic studies in populations with diverse ancestry and provides foundational imputation and allele frequency resources for future genetic studies in Mexico and in the United States where the Hispanic/Latino population is predominantly of Mexican descent.

genetics↗

bioRxiv2021

Within-sibship GWAS improve estimates of direct genetic effects

Estimates from genome-wide association studies (GWAS) represent a combination of the effect of inherited genetic variation (direct effects), demography (population stratification, assortative mating) and genetic nurture from relatives (indirect genetic effects). GWAS using family-based designs can control for demography and indirect genetic effects, but large-scale family datasets have been lacking. We combined data on 159,701 siblings from 17 cohorts to generate population (between-family) and within-sibship (within-family) estimates of genome-wide genetic associations for 25 phenotypes. We demonstrate that existing GWAS associations for height, educational attainment, smoking, depressive symptoms, age at first birth and cognitive ability overestimate direct effects. We show that estimates of SNP-heritability, genetic correlations and Mendelian randomization involving these phenotypes substantially differ when calculated using within-sibship estimates. For example, genetic correlations between educational attainment and height largely disappear. In contrast, analyses of most clinical phenotypes (e.g. LDL-cholesterol) were generally consistent between population and within-sibship models. We also report compelling evidence of polygenic adaptation on taller human height using within-sibship data. Large-scale family datasets provide new opportunities to quantify direct effects of genetic variation on human traits and diseases.

genetics↗

bioRxiv2020

Exome-wide association studies in general and long-lived populations identify genetic variants related to human age

Aging is characterized by degeneration in cellular and organismal functions leading to increased disease susceptibility and death. Although our understanding of aging biology in model systems has increased dramatically, large-scale sequencing studies to understand human aging are now just beginning. We applied exome sequencing and association analyses (ExWAS) to identify age-related variants on 58,470 participants of the DiscovEHR cohort. Linear Mixed Model regression analyses of age at last encounter revealed variants in genes known to be linked with clonal hematopoiesis of indeterminate potential, which are associated with myelodysplastic syndromes, as top signals in our analysis, suggestive of age-related somatic mutation accumulation in hematopoietic cells despite patients lacking clinical diagnoses. In addition to APOE, we identified rare DISP2 rs183775254 (p = 7.40x10-10) and ZYG11A rs74227999 (p = 2.50x10-08) variants that were negatively associated with age in either both sexes combined and females, respectively, which were replicated with directional consistency in two independent cohorts. Epigenetic mapping showed these variants are located within cell-type-specific enhancers, suggestive of important transcriptional regulatory functions. To discover variants associated with extreme age, we performed exome-sequencing on persons of Ashkenazi Jewish descent ascertained for extensive lifespans. Case-Control analyses in 525 Ashkenazi Jews cases (Males [≥] 92 years, Females [≥] 95years) were compared to 482 controls. Our results showed variants in APOE (rs429358, rs6857), and TMTC2 (rs7976168) passed Bonferroni-adjusted p-value, as well as several nominally-associated population-specific variants. Collectively, our Age-ExWAS, the largest performed to date, confirmed and identified previously unreported candidate variants associated with human age.

genetics↗

bioRxiv2018

Loss of ZnT8 function protects against diabetes by enhanced insulin secretion

A rare loss-of-function variant p.Arg138* in SLC30A8 encoding the zinc transporter 8 (ZnT8) enriched in Western Finland protects against type 2 diabetes (T2D). We recruited relatives of the identified carriers and showed that protection was associated with better insulin secretion due to enhanced glucose responsiveness and proinsulin conversion, especially compared with individuals matched for the genotype of a common T2D risk variant in SLC30A8, p.Arg325. In genome-edited human IPS-derived {beta}-like cells, we establish that the p.Arg138* variant results in reduced SLC30A8 expression due to haploinsufficiency. In human {beta}-cells loss of SLC30A8 leads to increased glucose responsiveness and reduced KATP channel function, which was also seen in isolated islets from carriers of the T2D-protective allele p.Trp325. These data position ZnT8 as an appealing target for treatment aiming at maintaining insulin secretion capacity in T2D.

genetics↗

bioRxiv2018

Convolving Pre-Trained Convolutional Neural Networks at Various Magnifications to Extract Diagnostic Features for Digital Pathology

Deep learning is an area of artificial intelligence that has received much attention in the past few years due to both an increase in computational power with the increased use of graphics processing units (GPUs) for computational analyses and the performance of these class of algorithms on visual recognition tasks. They have found utility in applications ranging from image search to facial recognition for security and social media purposes. Their continued success has propelled their use across many new domains including the medical field, in areas of radiology and pathology in particular, as these fields are thought to be driven by visual recognition tasks. In this paper, we present an application of deep learning, termed transfer learning, using ResNet50, a pre-trained convolutional neural network (CNN) to act as a feature-detector at various magnifications to identify low and high level features in digital pathology images of various breast lesions for the purpose of classifying them correctly into the labels of normal, benign, in-situ, or invasive carcinoma as provided in the ICIAR 2018 Breast Cancer Histology Challenge (BACH).

bioinformatics↗

bioRxiv2018

AI-MHC: an allele-integrated deep learning framework for improving Class I & Class II HLA-binding predictions

MotivationThe immune system has potential to present a wide variety of peptides to itself as a means of surveillance for pathogenic invaders. This means of surveillances allows the immune system to detect peptides derives from bacterial, viral, and even oncologic sources. However, given the breadth of the epitope repertoire, in order to study immune responses to these epitopes, investigators have relied on in-silico prediction algorithms to help narrow down the list of candidate epitopes, and current methods still have much in the way of improvement.\n\nResultsWe present Allele-Integrated MHC (AI-MHC), a deep learning architecture with improved performance over the current state-of-the-art algorithms in human Class I and Class II MHC binding prediction. Our architecture utilizes a convolutional neural network that improves prediction accuracy by 1) allowing one neural network to be trained on all peptides for all alleles of a given class of MHC molecules by making the allele an input to the net and 2) introducing a global max pooling operation with an optimized kernel size that allows the architecture to achieve translational invariance in MHC-peptide binding analysis, making it suitable for sequence analytics where a frame of interest needs to be learned in a longer, variable length sequence. We assess AI-MHC against internal independent test sets and compare against all algorithms in the IEDB automated server benchmarks, demonstrating our algorithm achieves state-of-the-art for both Class I and Class II prediction.\n\nAvailability and ImplementationAI-MHC can be used via web interface at baras.pathology.jhu.edu/AI-MHC\n\nContactjsidhom1@jhmi.edu

bioinformatics↗

bioRxiv2018

Genome-wide association study of 1 million people identifies 111 loci for atrial fibrillation

To understand the genetic variation underlying atrial fibrillation (AF), the most common cardiac arrhythmia, we performed a genome-wide association study (GWAS) of > 1 million people, including 60,620 AF cases and 970,216 controls. We identified 163 independent risk variants at 111 loci and prioritized 165 candidate genes likely to be involved in AF. Many of the identified risk variants fall near genes where more deleterious mutations have been reported to cause serious heart defects in humans or mice (MYH6, NKX2-5, PITX2, TBC1D32, TBX5),1,2 or near genes important for striated muscle function and integrity (e.g. MYH7, PKP2, SSPN, SGCA). Experiments in rabbits with heart failure and left atrial dilation identified a heterogeneous distributed molecular switch from MYH6 to MYH7 in the left atrium, which resulted in contractile and functional heterogeneity and may predispose to initiation and maintenance of atrial arrhythmia.

genetics↗

bioRxiv2017

KaryoScan: abnormal karyotype detection from whole-exome sequence

MotivationDetection of abnormal karyotypes from whole-exome sequencing has significant clinical potential, enabling a primary screen for chromosomal anomalies among samples undergoing short-read sequencing for nucleotide resolution genomic characterization.\n\nResultsWe present KaryoScan, a high-throughput method for detecting chromosomal anomalies within large cohort exome sequencing studies. We detect and validate autosomal and sex chromosomal aneuploidies in a large exome sequencing cohort, and demonstrate detection of smaller and complex events (partial chromosome, mosaic, copy neutral, and complex rearrangements), representing the range of anomalies that can be uncovered from the exome.\n\nAvailabilityhttps://github.com/rgcgithub/karyoscan

bioinformatics↗

bioRxiv2017

Profiling and leveraging relatedness in a precision medicine cohort of 92,455 exomes

Large-scale human genetics studies are ascertaining increasing proportions of populations as they continue growing in both number and scale. As a result, the amount of cryptic relatedness within these study cohorts is growing rapidly and has significant implications on downstream analyses. We demonstrate this growth empirically among the first 92,455 exomes from the DiscovEHR cohort and, via a custom simulation framework we developed called SimProgeny, show that these measures are in-line with expectations given the underlying population and ascertainment approach. For example, we identified [~]66,000 close (first- and second-degree) relationships within DiscovEHR involving 55.6% of study participants. Our simulation results project that >70% of the cohort will be involved in these close relationships as DiscovEHR scales to 250,000 recruited individuals. We reconstructed 12,574 pedigrees using these relationships (including 2,192 nuclear families) and leveraged them for multiple applications. The pedigrees substantially improved the phasing accuracy of 20,947 rare, deleterious compound heterozygous mutations. Reconstructed nuclear families were critical for identifying 3,415 de novo mutations in [~]1,783 genes. Finally, we demonstrate the segregation of known and suspected disease-causing mutations through reconstructed pedigrees, including a tandem duplication in LDLR causing familial hypercholesterolemia. In summary, this work highlights the prevalence of cryptic relatedness expected among large healthcare population genomic studies and demonstrates several analyses that are uniquely enabled by large amounts of cryptic relatedness.

genomics↗

bioRxiv2017

Profiling copy number variation and disease associations from 50,726 DiscovEHR Study exomes

Copy number variants (CNVs) are a substantial source of genomic variation and contribute to a wide range of human disorders. Gene-disrupting exonic CNVs have important clinical implications as they can underlie variability in disease presentation and susceptibility. The relationship between exonic CNVs and clinical traits has not been broadly explored at the population level, primarily due to technical challenges. We surveyed common and rare CNVs in the exome sequences of 50,726 adult DiscovEHR study participants with linked electronic health records (EHRs). We evaluated the diagnostic yield and clinical expressivity of known pathogenic CNVs, and performed tests of association with EHR-derived serum lipids, thereby evaluating the relationship between CNVs and complex traits and phenotypes in an unbiased, real-world clinical context. We identified CNVs from megabase to exon-level resolution, demonstrating reliable, high-throughput detection of clinically relevant exonic CNVs. In doing so, we created a catalog of high-confidence common and rare CNVs and refined population frequency estimates of known and novel gene-disrupting CNVs. Our survey among an unselected clinical population provides further evidence that neuropathy-associated duplications and deletions in 17p12 have similar population prevalence but are clinically under-diagnosed. Similarly, adults who harbor 22q11.2 deletions frequently had EHR documentation of neurodevelopmental/neuropsychiatric disorders and congenital anomalies, but not a formal genetic diagnosis (i.e., deletion). In an exome-wide association study of lipid levels, we identified a novel five-exon duplication within LDLR segregating in a large kindred with features of familial hypercholesterolemia. Exonic CNVs provide new opportunities to understand and diagnose human disease.

genomics↗

bioRxiv2017

Mutation spectrum of NOD2 reveals recessive inheritance as a main driver of Early Onset Crohn’s Disease

Inflammatory bowel disease (IBD), clinically defined as Crohns disease (CD), ulcerative colitis (UC), or IBD-unclassified, results in chronic inflammation of the gastrointestinal tract in genetically susceptible hosts. Pediatric onset IBD represents [≥]25% of all IBD diagnoses and often presents with intestinal stricturing, perianal disease, and failed response to conventional treatments. NOD2 was the first and is the most replicated locus associated with adult IBD, to date. To determine the role of NOD2 and other genes in pediatric IBD, we performed whole-exome sequencing on a cohort of 1,183 patients with pediatric onset IBD (ages 0-18.5 years). We identified 92 probands who were homozygous or compound heterozygous for rare and low frequency NOD2 variants accounting for approximately 8% of our cohort, suggesting a Mendelian recessive inheritance pattern of disease. Additionally, we investigated the contribution of recessive inheritance of NOD2 alleles in adult IBD patients from the Regeneron Genetics Center (RGC)-Geisinger Health System DiscovEHR study, which links whole exome sequences to longitudinal electronic health records (EHRs) from 51,289 participants. We found that ~7% of cases in this adult IBD cohort, including ~10% of CD cases, can be attributed to recessive inheritance of NOD2 variants, confirming the observations from our pediatric IBD cohort. Exploration of EHR data showed that 14% of these adult IBD patients obtained their initial IBD diagnosis before 18 years of age, consistent with early onset disease. Collectively, our findings show that recessive inheritance of rare and low frequency deleterious NOD2 variants account for 7-10% of CD cases and implicate NOD2 as a Mendelian disease gene for early onset Crohns Disease.\n\nAuthor SummaryPediatric onset inflammatory bowel disease (IBD) represents [≥]25% of IBD diagnoses; yet the genetic architecture of early onset IBD remains largely uncharacterized. To investigate this, we performed whole-exome sequencing and rare variant analysis on a cohort of 1,183 pediatric onset IBD patients. We found that 8% of patients in our cohort were homozygous or compound heterozygous for rare or low frequency deleterious variants in the nucleotide binding and oligomerization domain containing 2 (NOD2) gene. Further investigation of whole-exome sequencing of a large clinical cohort of adult IBD patients uncovered recessive inheritance of rare and low frequency NOD2 variants in 7% of cases and that the relative risk for NOD2 variant homozygosity has likely been underestimated. While it has been reported that having >1 NOD2 risk alleles is associated with increased susceptibility to Crohns Disease (CD), our data formally demonstrate what has long been suspected: recessive inheritance of NOD2 alleles is a mechanistic driver of early onset IBD, specifically CD, likely due to loss of NOD2 protein function. Our data suggest that a subset of IBD-CD patients with early disease onset is characterized by recessive inheritance of NOD2 alleles, which has important implications for the screening, diagnosis, and treatment of IBD.

genetics↗