Biology subjects

Song, Y. S.

Publications and source records attributed to Song, Y. S..

At least 37 records · Page 2Linked to original sources

bioRxiv2024

Highly parameterized polygenic scores tend to overfit to population stratification via random effects

Polygenic scores (PGSs), increasingly used in clinical settings, frequently include many genetic variants, with performance typically peaking at thousands of variants. Such highly parameterized PGSs often include variants that do not pass a genome-wide significance threshold. We propose a mathematical perspective that renders the effects of many of these nonsignificant variants random rather than causal, with the randomness capturing population structure. We devise methods to assess variant effect randomness and population stratification bias. Applying these methods to 141 traits from the UK Biobank, we find that, for many PGSs, the effects of non-significant variants are considerably random, with the extent of randomness associated with the degree of overfitting to population structure of the discovery cohort. Our findings explain why highly parameterized PGSs simultaneously have superior cohort-specific performance and limited generalizability, suggesting the critical need for variant randomness tests in PGS evaluation. Supporting code and a dashboard are available at https://github.com/songlab-cal/StratPGS.

genetics↗

bioRxiv2023

ConvexML: Scalable and accurate inference of single-cell chronograms from CRISPR/Cas9 lineage tracing data

Branch length estimation is a fundamental problem in Statistical Phylogenetics and a core component of tree reconstruction algorithms. Traditionally, general time-reversible mutation models are employed, and many software tools exist for this scenario. With the advent of CRISPR/Cas9 lineage tracing technologies, there has been significant interest in the study of branch length estimation under irreversible mutation models. Under the CRISPR/Cas9 mutation model, irreversible mutations - in the form of DNA insertions or deletions - are accrued during the experiment, which are then read out at the single-cell level to reconstruct the cell lineage tree. However, most of the analyses of CRISPR/Cas9 lineage tracing data have so far been limited to the reconstruction of single-cell tree topologies, which depict lineage relationships between cells, but not the amount of time that has passed between ancestral cell states and the present. Time-resolved trees, known as chronograms, would allow one to study the evolutionary dynamics of cell populations at an unprecedented level of resolution. Indeed, time-resolved trees would reveal the timing of events on the tree, the relative fitness of subclones, and the dynamics underlying phenotypic changes in the cell population - among other important applications. In this work, we introduce the first scalable and accurate method to refine any given single-cell tree topology into a single-cell chronogram by estimating its branch lengths. To do this, we perform regularized maximum likelihood estimation under a general irreversible mutation model, paired with a conservative version of maximum parsimony that reconstructs only the ancestral states that we are confident about. To deal with the particularities of CRISPR/Cas9 lineage tracing data - such as double-resection events which affect runs of consecutive sites - we avoid making our model more complex and instead opt for using a simple but effective data encoding scheme. Similarly, we avoid explicitly modeling the missing data mechanisms - such as heritable missing data - by instead assuming that they are missing completely at random. We stabilize estimates in low-information regimes by using a simple penalized version of maximum likelihood estimation (MLE) using a minimum branch length constraint and pseudocounts. All this leads to a convex MLE problem that can be readily solved in seconds with off-the-shelf convex optimization solvers. We benchmark our method using both simulations and real lineage tracing data, and show that it performs well on several tasks, matching or outperforming competing methods such as TiDeTree and LAML in terms of accuracy, while being 10 ~ 100 x faster. Notably, our statistical model is simpler and more general, as we do not explicitly model the intricacies of CRISPR/Cas9 lineage tracing data. In this sense, our contribution is twofold: (1) a fast and robust method for branch length estimation under a gen-eral irreversible mutation model, and (2) a data encoding scheme specific to CRISPR/Cas9-lineage tracing data which makes it amenable to the general model. Our branch length estimation method, which we call ConvexML, should be broadly applicable to any evolutionary model with irreversible mutations (ideally, with high diversity) and an approximately ignorable missing data mechanism. ConvexML is available through the convexml open source Python package.

bioinformatics↗

bioRxiv2023

GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction

Whereas protein language models have demonstrated remarkable efficacy in predicting the effects of missense variants, DNA counterparts have not yet achieved a similar competitive edge for genome-wide variant effect predictions, especially in complex genomes such as that of humans. To address this challenge, we here introduce GPN-MSA, a novel framework for DNA language models that leverages whole-genome sequence alignments across multiple species and takes only a few hours to train. Across several benchmarks on clinical databases (ClinVar, COSMIC, OMIM), experimental functional assays (DMS, DepMap), and population genomic data (gnomAD), our model for the human genome achieves outstanding performance on deleteriousness prediction for both coding and non-coding variants.

bioinformatics↗

bioRxiv2023

CELL-E 2: Translating Proteins to Pictures and Back with a Bidirectional Text-to-Image Transformer

We present CELL-E 2, a novel bidirectional transformer that can generate images depicting protein subcellular localization from the amino acid sequences (and vice versa). Protein localization is a challenging problem that requires integrating sequence and image information, which most existing methods ignore. CELL-E 2 extends the work of CELL-E, not only capturing the spatial complexity of protein localization and produce probability estimates of localization atop a nucleus image, but also being able to generate sequences from images, enabling de novo protein design. We train and finetune CELL-E 2 on two large-scale datasets of human proteins. We also demonstrate how to use CELL-E 2 to create hundreds of novel nuclear localization signals (NLS). Results and interactive demos are featured at https://bohuanglab.github.io/CELL-E_2/.

bioinformatics↗

bioRxiv2023

Predicting the effect of CRISPR-Cas9-based epigenome editing

Epigenetic regulation orchestrates mammalian transcription, but functional links between them remain elusive. To tackle this problem, we use epigenomic and transcriptomic data from 13 ENCODE cell types to train machine learning models to predict gene expression from histone post-translational modifications (PTMs), achieving transcriptome-wide correlations of [~] 0.70 - 0.79 for most cell types. Our models recapitulate known associations between histone PTMs and expression patterns, including predicting that acetylation of histone subunit H3 lysine residue 27 (H3K27ac) near the transcription start site (TSS) significantly increases expression levels. To validate this prediction experimentally and investigate how natural vs. engineered deposition of H3K27ac might differentially affect expression, we apply the synthetic dCas9-p300 histone acetyltransferase system to 8 genes in the HEK293T cell line and to 5 genes in the K562 cell line. Further, to facilitate model building, we perform MNase-seq to map genome-wide nucleosome occupancy levels in HEK293T. We observe that our models perform well in accurately ranking relative fold-changes among genes in response to the dCas9-p300 system; however, their ability to rank fold-changes within individual genes is noticeably diminished compared to predicting expression across cell types from their native epigenetic signatures. Our findings highlight the need for more comprehensive genome-scale epigenome editing datasets, better understanding of the actual modifications made by epigenome editing tools, and improved causal models that transfer better from endogenous cellular measurements to perturbation experiments. Together these improvements would facilitate the ability to understand and predictably control the dynamic human epigenome with consequences for human health.

bioinformatics↗

bioRxiv2023

Reference-free multiplexed single-cell sequencing identifies genetic modifiers of the human immune response

Multiplexed single-cell sequencing (mux-seq) using single-nucleotide polymorphisms (SNPs) has emerged as an efficient approach to perform expression quantitative trait loci (eQTL) studies that map interactions between genetic variants and cell types, cell states, or experimental perturbations. Here we introduce the clue framework, a novel approach to encode mux-seq experiments that eliminates the need for reference genotypes and experimental barcoding. The clue framework is made possible by the development of freemuxlet, an algorithm that clusters cells based on SNPs called from single-cell RNA-seq or ATAC-seq data. To demonstrate the feasibility of clue, we profiled the surface protein and RNA abundances of peripheral blood mononuclear cells from 64 individuals, stimulated with 5 distinct extracellular stimuli -- all within a single day. Our analysis of the demultiplexed data identified rare immune cell types and cell type-specific responses to interferon and toll-like receptor stimulation. Furthermore, by integrating genotyping data, we mapped response eQTLs specific to certain cell types. These findings showcase the potential and scalability of the clue framework for reference-free multiplexed single-cell sequencing studies.

genomics↗

bioRxiv2023

The Impact of Stability Considerations on Genetic Fine-Mapping

Fine-mapping methods, which aim to identify genetic variants responsible for complex traits following genetic association studies, typically assume that sufficient adjustments for confounding within the association study cohort have been made, e.g., through regressing out the top principal components (i.e., residualization). Despite its widespread use, however, residualization may not completely remove all sources of confounding. Here, we propose a complementary stability-guided approach that does not rely on residualization, which identifies consistently fine-mapped variants across different genetic backgrounds or environments. Simulations show that stability guidance neither outperforms nor underperforms residualization, but each approach picks up different variants considerably often. Critically, prioritizing variants that match between the residualization and stability-guided approaches enhances recovery of causal variants. We further demonstrate the utility of the stability approach by applying it to fine-map eQTLs in the GEUVADIS data. Using 378 different functional annotations of the human genome, including recent deep learning-based annotations (e.g., Enformer), we compare enrichments of these annotations among variants for which the stability and traditional residualization-based fine-mapping approaches agree against those for which they disagree, and find that the stability approach enhances the power of traditional fine-mapping methods in identifying variants with functional impact. Finally, in cases where the two approaches report distinct variants, our approach identifies variants comparably enriched for functional annotations. Our findings suggest that the stability principle, as a conceptually simple device, complements existing approaches to fine-mapping, reinforcing recent advocacy of evaluating cross-population and cross-environment portability of biological findings. To support visualization and interpretation of our results, we provide a Shiny app, available at: https://alan-aw.shinyapps.io/stability_v0/.

genetics↗

bioRxiv2023

Tracing cancer evolution and heterogeneity using Hi-C

Chromosomal rearrangements can initiate and drive cancer progression, yet it has been challenging to evaluate their impact, especially in genetically heterogeneous solid cancers. To address this problem we developed HiDENSEC, a new computational framework for analyzing chromatin conformation capture in heterogeneous samples, which can infer somatic copy number alterations, characterize large-scale chromosomal rearrangements, and estimate cancer cell fractions. We validated HiDENSEC with in silico and in vitro controls, and then characterized chromosome-scale evolution during melanoma progression in formalin-fixed tumor samples from three patients. The resulting comprehensive annotation of the genomic events includes copy number neutral translocations that disrupt tumor suppressor genes such as NF1, whole chromosome arm exchanges that result in loss of CDKN2A, and whole-arm copy-number neutral loss of homozygosity involving PTEN. These findings show that large-scale chromosomal rearrangements occur throughout cancer evolution and characterizing these events yields insights into drivers of melanoma progression.

genomics↗

bioRxiv2022

CherryML: Scalable Maximum Likelihood Estimation of Phylogenetic Models

Phylogenetic models of molecular evolution are central to diverse problems in biology, but maximum likelihood estimation of model parameters is a computationally expensive task, in some cases prohibitively so. To address this challenge, we here introduce CherryML, a broadly applicable method that achieves several orders of magnitude speedup. We demonstrate its utility by applying it to estimate a general 400 x 400 rate matrix for amino acid co-evolution at protein contact sites.

evolutionary biology↗

bioRxiv2022

Cross-protein transfer learning substantially improves zero-shot prediction of disease variant effects

Genetic variation in the human genome is a major determinant of individual disease risk, but the vast majority of missense variants have unknown etiological effects. Here, we present a robust learning framework for leveraging saturation mutagenesis experiments to construct accurate computational predictors of proteome-wide missense variant pathogenicity. We train cross-protein transfer (CPT) models using deep mutational scanning data from only five proteins and achieve state-of-the-art performance on clinical variant interpretation for unseen proteins across the human proteome. High sensitivity is crucial for clinical applications and our model CPT-1 particularly excels in this regime. For instance, at 95% sensitivity of detecting human disease variants annotated in ClinVar, CPT-1 improves specificity to 68%, from 27% for ESM-1v and 55% for EVE. Furthermore, for genes not used to train REVEL, a supervised method widely used by clinicians, we show that CPT-1 compares favorably with REVEL. Our framework combines predictive features derived from general protein sequence models, vertebrate sequence alignments, and AlphaFold2 structures, and it is adaptable to the future inclusion of other sources of information. We find that vertebrate alignments, albeit rather shallow with only 100 genomes, provide a strong signal for variant pathogenicity prediction that is complementary to recent deep learning-based models trained on massive amounts of protein sequence data. We release predictions for all possible missense variants in 90% of human genes. Our results demonstrate the utility of mutational scanning data for learning properties of variants that transfer to unseen proteins.

genetics↗

bioRxiv2022

Tuned Fitness Landscapes for Benchmarking Model-Guided Protein Design

Advancements in DNA synthesis and sequencing technologies have enabled a novel paradigm of protein design where machine learning (ML) models trained on experimental data are used to guide exploration of a protein fitness landscape. ML-guided directed evolution (MLDE) builds on the success of traditional directed evolution and unlocks strategies which make more efficient use of experimental data. Building an MLDE pipeline involves many design choices across the design-build-test-learn loop ranging from data collection strategies to modeling, each of which has a large impact on the success of designed sequences. The cost of collecting experimental data makes benchmarking every component of these pipelines on real data prohibitively difficult, necessitating the development of synthetic landscapes where MLDE strategies can be tested. In this work, we develop a framework called SLIP ("Synthetic Landscape Inference for Proteins") for constructing biologically-motivated synthetic landscapes with tunable difficulty based on Potts models. This framework can be extended to any protein family for which there is a sequence alignment. We show that without tuning, Potts models are easy to optimize. In contrast, our tuning framework provides landscapes sufficiently challenging to benchmark MLDE pipelines. SLIP is open-source and is available at https://github.com/google-research/slip.

synthetic biology↗

bioRxiv2022

The Impact of Protein Dynamics on Residue-Residue Coevolution and Contact Prediction

The need to maintain protein structure constrains evolution at the sequence level, and patterns of coevolution in homologous protein sequences can be used to predict their 3D structures with high accuracy. Our understanding of the relationship between protein structure and evolution has traditionally been benchmarked by computational models ability to predict contacts from a single representative, experimentally determined structure per protein family. However, proteins in vivo are highly dynamic and can adopt multiple functionally relevant conformations. Here we demonstrate that interactions that stabilize alternate conformations, as well those that mediate conformational changes, impose an underappreciated but significant set of evolutionary constraints. We analyze the extent of these constraints over 56 paralogous G protein coupled receptors (GPCRs), {beta}-arrestin and the human SARS-CoV2 receptor ACE2. Specifically, we observe that contacts uniquely found in molecular dynamics (MD) simulation data and alternate-conformation crystal structures are successfully predicted by unsupervised language models. In GPCRs, adding these contacts as positives increases the percentage of top contacts classified as true positives, as predicted by a state-of-the-art language model, from 69% to 87%. Our results show that protein dynamics impose constraints on molecular evolution and demonstrate the ability of unsupervised language models to measure these constraints.

bioinformatics↗

bioRxiv2022

DNA language models are powerful zero-shot predictors of non-coding variant effects

The expanding catalog of genome-wide association studies (GWAS) provides biological insights across a variety of species, but identifying the causal variants behind these associations remains a significant challenge. Experimental validation is both labor-intensive and costly, highlighting the need for accurate, scalable computational methods to predict the effects of genetic variants across the entire genome. Inspired by recent progress in natural language processing, unsupervised pre-training on large protein sequence databases has proven successful in extracting complex information related to proteins. These models showcase their ability to learn variant effects in coding regions using an unsupervised approach. Expanding on this idea, we here introduce the Genomic Pre-trained Network (GPN), a model designed to learn genome-wide variant effects through unsupervised pre-training on genomic DNA sequences. Our model also successfully learns gene structure and DNA motifs without any supervision. To demonstrate its utility, we train GPN on unaligned reference genomes of Arabidopsis thaliana and seven related species within the Brassicales order, and evaluate its ability to predict the functional impact of genetic variants in Arabidopsis thaliana by utilizing allele frequencies from the 1001 Genomes Project and a comprehensive database of GWAS. Notably, GPN outperforms predictors based on popular conservation scores such as phyloP and phastCons. Our predictions for Arabidopsis thaliana can be visualized as sequence logos in the UCSC Genome Browser (https://genome.ucsc.edu/s/gbenegas/gpn-arabidopsis). We provide code (https://github.com/songlab-cal/gpn) to train GPN for any given species using its DNA sequence alone, enabling unsupervised prediction of variant effects across the entire genome.

bioinformatics↗

bioRxiv2022

The ENCODE Imputation Challenge: A critical assessment of methods for cross-cell type imputation of epigenomic profiles

Functional genomics experiments are invaluable for understanding mechanisms of gene regulation. However, comprehensively performing all such experiments, even across a fixed set of sample and assay types, is often infeasible in practice. A promising alternative to performing experiments exhaustively is to, instead, perform a core set of experiments and subsequently use machine learning methods to impute the remaining experiments. However, questions remain as to the quality of the imputations, the best approaches for performing imputations, and even what performance measures meaningfully evaluate performance of such models. In this work, we address these questions by comprehensively analyzing imputations from 23 imputation models submitted to the ENCODE Imputation Challenge. We find that measuring the quality of imputations is significantly more challenging than reported in the literature, and is confounded by three factors: major distributional shifts that arise because of differences in data collection and processing over time, the amount of available data per cell type, and redundancy among performance measures. Our systematic analyses suggest several steps that are necessary, but also simple, for fairly evaluating the performance of such models, as well as promising directions for more robust research in this area.

bioinformatics↗

bioRxiv2022

CELL-E: Biological zero-shot text-to-image synthesis for protein localization prediction

Accurately predicting cellular activities of proteins based on their primary amino acid sequences would greatly improve our understanding of the proteome. In this paper, we present CELL-E, a text-to-image transformer architecture that generates a 2D probability density map of protein distribution within cells. Given a amino acid sequence and a reference image for cell or nucleus morphology, CELL-E offers a more direct representation of protein localization, as opposed to previous in silico methods that rely on pre-defined, discrete class annotations of protein localization to subcellular compartments.

bioinformatics↗

bioRxiv2022

A fast machine-learning-guided primer design pipeline for selective whole genome amplification

Addressing many of the major outstanding questions in the fields of microbial evolution and pathogenesis will require analyses of populations of microbial genomes. Although population genomic studies provide the analytical resolution to investigate evolutionary and mechanistic processes at fine spatial and temporal scales - precisely the scales at which these processes occur - microbial population genomic research is currently hindered by the practicalities of obtaining sufficient quantities of the relatively pure microbial genomic DNA necessary for next-generation sequencing. Here we present swga2.0, an optimized and parallelized pipeline to design selective whole genome amplification (SWGA) primer sets. Unlike previous methods, swga2.0 incorporates active and machine learning methods to evaluate the amplification efficacy of individual primers and primer sets. Additionally, swga2.0 optimizes primer set search and evaluate strategies, including parallelization at each stage of the pipeline, to dramatically decrease program runtime from weeks to minutes. Here we describe the swga2.0 pipeline, including the empirical data used to identify primer and primer set characteristics, that improve amplification performance. Additionally, we evaluated the novel swga2.0 pipeline by designing primers sets that successfully amplify Prevotella melaninogenica, an important component of the lung microbiome in cystic fibrosis patients, from samples dominated by human DNA.

genomics↗

bioRxiv2021

Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions

There is significant interest in developing machine learning methods to model protein-ligand interactions but a scarcity of experimentally resolved protein-ligand structures to learn from. Protein self-contacts are a much larger source of structural data that could be leveraged, but currently it is not well understood how this data source differs from the target domain. Here, we characterize the 3D geometric patterns of protein self-contacts as probability distributions. We then present a flexible statistical framework to assess the transferability of these patterns to protein-ligand contacts. We observe that the level of transferability from protein self-contacts to protein-ligand contacts depends on contact type, with many contact types exhibiting high transferability. We then demonstrate the potential of leveraging information from these geometric patterns to aid in ligand pose-selection problems in protein-ligand docking. We publicly release our extracted data on geometric interaction patterns to enable further exploration of this problem.

biophysics↗

bioRxiv2021

Robust and annotation-free analysis of isoform variation using short-read scRNA-seq data

Although alternative splicing is a fundamental and pervasive aspect of gene expression in higher eukaryotes, it is often omitted from single-cell studies due to quantification challenges inherent to commonly used short-read sequencing technologies. Here, we undertake the analysis of alternative splicing across numerous diverse murine cell types from two large-scale single-cell datasets--the Tabula Muris and BRAIN Initiative Cell Census Network--while accounting for understudied technical artifacts and unannotated isoforms. We find strong and general cell-type-specific alternative splicing, complementary to total gene expression but of similar discriminatory value, and identify a large volume of novel isoforms. We specifically highlight splicing variation across different cell types in primary motor cortex neurons, bone marrow B cells, and various epithelial cells; and show that the implicated transcripts include many genes which do not display total expression differences. To elucidate the regulation of alternative splicing, we build a custom predictive model based on splicing factor activity, recovering several known interactions while generating new hypotheses, including potential regulatory roles for novel alternative splicing events in critical genes including Khdrbs3 and Rbfox1. We make our results available using public interactive browsers to spur further exploration by the community.

genomics↗