Biology subjects

Kundaje, A.

Publications and source records attributed to Kundaje, A..

At least 91 records · Page 5Linked to original sources

bioRxiv2020

Chromatin and gene-regulatory dynamics of the developing human cerebral cortex at single-cell resolution

Genetic perturbations of cerebral cortical development can lead to neurodevelopmental disease, including autism spectrum disorder (ASD). To identify genomic regions crucial to corticogenesis, we mapped the activity of gene-regulatory elements generating a single-cell atlas of gene expression and chromatin accessibility both independently and jointly. This revealed waves of gene regulation by key transcription factors (TFs) across a nearly continuous differentiation trajectory into glutamatergic neurons, distinguished the expression programs of glial lineages, and identified lineage-determining TFs that exhibited strong correlation between linked gene-regulatory elements and expression levels. These highly connected genes adopted an active chromatin state in early differentiating cells, consistent with lineage commitment. Basepair-resolution neural network models identified strong cell-type specific enrichment of noncoding mutations predicted to be disruptive in a cohort of ASD subjects and identified frequently disrupted TF binding sites. This approach illustrates how cell-type specific mapping can provide insights into the programs governing human development and disease.

neuroscience↗

bioRxiv2020

Benchmarking Reverse-Complement Strategies for Deep Learning Models in Genomics

Predictive models mapping double-stranded DNA to signals of regulatory activity should, in principle, produce analogous (or "equivariant") predictions whether the forward strand or its reverse complement (RC) is supplied as input. Unfortunately, standard neural networks can produce highly divergent predictions across strands, even when the training set is augmented with RC sequences. Two strategies have emerged to enforce equivariance: conjoined/"siamese" architectures, and RC parameter sharing or RCPS. However, the connections between the two remain unclear, comparisons to strong baselines are lacking, and neither has been adapted to base-resolution signal profile prediction. Here we extend conjoined & RCPS models to base-resolution signal prediction, and introduce a strong baseline: a standard model (trained with RC data augmentation) that is made conjoined only after training, which we call "post-hoc" conjoined. Through benchmarks on diverse tasks, we find post-hoc conjoined consistently performs best or second-best, surpassed only occasionally by RCPS, and never underperforms conjoined-during-training. We propose an overfitting-based hypothesis for the latter finding, and study it empirically. Despite its theoretical appeal, RCPS shows mediocre performance on several tasks, even though (as we prove) it can represent any solution learned by conjoined models. Our results suggest users interested in RC equivariance should default to post-hoc conjoined as a reliable baseline before exploring RCPS. Finally, we present a unified description of conjoined & RCPS architectures, revealing a broader class of models that gradually interpolate between RCPS and conjoined while maintaining equivariance.

bioinformatics↗

bioRxiv2020

The dynamic, combinatorial cis-regulatory lexicon of epidermal differentiation

Transcription factors (TFs) bind DNA sequence motif vocabularies in cis-regulatory elements (CREs) to modulate chromatin state and gene expression during cell state transitions. A quantitative understanding of how motif lexicons influence dynamic regulatory activity has been elusive due to the combinatorial nature of the cis-regulatory code. To address this, we undertook multi-omic data profiling of chromatin and expression dynamics across epidermal differentiation to identify 40,103 dynamic CREs associated with 3,609 dynamically expressed genes, then applied an interpretable deep learning framework to model the cis-regulatory logic of chromatin accessibility. This identified cooperative DNA sequence rules in dynamic CREs regulating synchronous gene modules with diverse roles in skin differentiation. Massively parallel reporter analysis validated temporal dynamics and cooperative cis-regulatory logic. Variants linked to human polygenic skin disease were enriched in these time-dependent combinatorial motif rules. This integrative approach reveals the combinatorial cis-regulatory lexicon of epidermal differentiation and represents a general framework for deciphering the organizational principles of the cis-regulatory code in dynamic gene regulation. HIGHLIGHTSO_LIAn integrative multi-omic resource profiling chromatin and expression dynamics across keratinocyte differentiation C_LIO_LIPredictive deep learning models of chromatin dynamics reveal a high-resolution cis-regulatory DNA motif lexicon of epidermal differentiation C_LIO_LIModel interpretation enables discovery of combinatorial cis-regulatory logic of homotypic and heterotypic motif combinations C_LIO_LIMassively parallel reporter experiments validate temporal dynamics and cis-regulatory logic of the combinatorial motif lexicon C_LI

genomics↗

bioRxiv2020

fastISM: Performant in-silico saturation mutagenesis for convolutional neural networks

Deep learning models such as convolutional neural networks are able to accurately map biological sequences to associated functional readouts and properties by learning predictive de novo representations. In-silico saturation mutagenesis (ISM) is a popular feature attribution technique for inferring contributions of all characters in an input sequence to the models predicted output. The main drawback of ISM is its runtime, as it involves multiple forward propagations of all possible mutations of each character in the input sequence through the trained model to predict the effects on the output. We present fastISM, an algorithm that speeds up ISM by a factor of over 10x for commonly used convolutional neural network architectures. fastISM is based on the observations that the majority of computation in ISM is spent in convolutional layers, and a single mutation only disrupts a limited region of intermediate layers, rendering most computation redundant. fastISM reduces the gap between backpropagation-based feature attribution methods and ISM. It far surpasses the runtime of backpropagation-based methods on multi-output architectures, making it feasible to run ISM on a large number of sequences. An easy-to-use Keras/TensorFlow 2 implementation of fastISM is available at https://github.com/kundajelab/fastISM, and a hands-on tutorial at https://colab.research.google.com/github/kundajelab/fastISM/blob/master/notebooks/colab/DeepSEA.ipynb.

bioinformatics↗

bioRxiv2020

High-throughput discovery and characterization of human transcriptional effectors

Thousands of proteins localize to the nucleus; however, it remains unclear which contain transcriptional effectors. Here, we develop HT-recruit - a pooled assay where protein libraries are recruited to a reporter, and their transcriptional effects are measured by sequencing. Using this approach, we measure gene silencing and activation for thousands of domains. We find a relationship between repressor function and evolutionary age for the KRAB domains, discover Homeodomain repressor strength is collinear with Hox genetic organization, and identify activities for several Domains of Unknown Function. Deep mutational scanning of the CRISPRi KRAB maps the co-repressor binding surface and identifies substitutions that improve stability/silencing. By tiling 238 proteins, we find repressors as short as 10 amino acids. Finally, we report new activator domains, including a divergent KRAB. Together, these results provide a resource of 600 human proteins containing effectors and demonstrate a scalable strategy for assigning functions to protein domains.

genetics↗

bioRxiv2020

Genome-wide maps of enhancer regulation connect risk variants to disease genes

Genome-wide association studies have now identified tens of thousands of noncoding loci associated with human diseases and complex traits, each of which could reveal insights into biological mechanisms of disease. Many of the underlying causal variants are thought to affect enhancers, but we have lacked genome-wide maps of enhancer-gene regulation to interpret such variants. We previously developed the Activity-by-Contact (ABC) Model to predict enhancer-gene connections and demonstrated that it can accurately predict the results of CRISPR perturbations across several cell types. Here, we apply this ABC Model to create enhancer-gene maps in 131 cell types and tissues, and use these maps to interpret the functions of fine-mapped GWAS variants. For inflammatory bowel disease (IBD), causal variants are >20-fold enriched in enhancers in particular cell types, and ABC outperforms other regulatory methods at connecting noncoding variants to target genes. Across 72 diseases and complex traits, ABC links 5,036 GWAS signals to 2,249 unique genes, including a class of 577 genes that appear to influence multiple phenotypes via variants in enhancers that act in different cell types. Guided by these variant-to-function maps, we show that an enhancer containing an IBD risk variant regulates the expression of PPIF to tune mitochondrial membrane potential. Together, our study reveals insights into principles of genome regulation, illuminates mechanisms that influence IBD, and demonstrates a generalizable strategy to connect common disease risk variants to their molecular and cellular functions.

genetics↗

bioRxiv2020

Size-independent mRNA synthesis and chromatin-based partitioning mechanisms generate and maintain constant amounts of protein per cell

Biosynthesis scales with cell size such that protein concentrations generally remain constant as cells grow. As an exception, synthesis of the cell-cycle inhibitor Whi5 sub-scales with cell size so that its concentration is lower in larger cells to promote cell-cycle entry. Here, we find that a transcriptional control uncouples Whi5 synthesis from cell size and, screening for similar genes, identify histones as the major class of sub-scaling transcripts besides WHI5. Histone synthesis is thereby matched to genome content rather than cell size. Such sub-scaling proteins are challenged by asymmetric cell division because proteins are typically partitioned in proportion to new-born cell volume. To avoid this fate, Whi5 uses chromatin-binding to partition similar protein amounts to each new-born cell regardless of cell size. Finally, disrupting both Whi5 synthesis and chromatin-based partitioning compromises G1 size control. Thus, specific transcriptional and partitioning mechanisms determine protein sub-scaling to control cell size.

cell biology↗

bioRxiv2020

Transcription-dependent domain-scale 3D genome organization in dinoflagellates

Dinoflagellate chromosomes represent a unique evolutionary experiment, as they exist in a permanently condensed, liquid crystalline state, are not packaged by histones, and contain genes organized into polycistronic arrays, with minimal transcriptional regulation. We analyze the 3D genome of Breviolum minutum, and find large topological domains without chromatin loops, demarcated by convergent gene array boundaries ("dinoTADs). Transcriptional inhibition degrades dinoTADs, implicating transcription-induced supercoiling as the primary topological force in dinoflagellates.

genomics↗

bioRxiv2020

Fourier-transform-based attribution priors improve the interpretability and stability of deep learning models for genomics

Deep learning models can accurately map genomic DNA sequences to associated functional molecular readouts such as protein-DNA binding data. Base-resolution importance (i.e. "attribution") scores inferred from these models can highlight predictive sequence motifs and syntax. Unfortunately, these models are prone to overfitting and are sensitive to random initializations, often resulting in noisy and irreproducible attributions that obfuscate underlying motifs. To address these shortcomings, we propose a novel attribution prior, where the Fourier transform of input-level attribution scores are computed at training-time, and high-frequency components of the Fourier spectrum are penalized. We evaluate different model architectures with and without attribution priors trained on genome-wide binary or continuous molecular profiles. We show that our attribution prior dramatically improves models stability, interpretability, and performance on held-out data, especially when training data is severely limited. Our attribution prior also allows models to identify biologically meaningful sequence motifs more sensitively and precisely within individual regulatory elements. The prior is agnostic to the model architecture or predicted experimental assay, yet provides similar gains across all experiments. This work represents an important advancement in improving the reliability of deep learning models for deciphering the regulatory code of the genome.

bioinformatics↗

bioRxiv2020

MTSplice predicts effects of genetic variants on tissue-specific splicing

Tissue-specific splicing of exons plays an important role in determining tissue identity. However, computational tools predicting tissue-specific effects of variants on splicing are lacking. To address this issue, we developed MTSplice (Multi-tissue Splicing), a neural network which quantitatively predicts effects of human genetic variants on splicing of cassette exons in 56 tissues. MTSplice combines the state-of-the-art predictor MMSplice, which models constitutive regulatory sequences, with a new neural network which models tissue-specific regulatory sequences. MTSplice outperforms MMSplice on predicting effects associated with naturally occurring genetic variants in most tissues of the GTEx dataset. Furthermore, MTSplice predicts that autism-associated de novo mutations are enriched for variants affecting splicing specifically in the brain. MTSplice is provided free of use and open source at the model repository Kipoi. We foresee MTSplice to be useful for functional prediction and prioritization of variants associated with tissue-specific disorders.

genomics↗

bioRxiv2020

Single-cell epigenomic identification of inherited risk loci in Alzheimer's and Parkinson's disease

Genome-wide association studies (GWAS) have identified thousands of variants associated with disease phenotypes. However, the majority of these variants do not alter coding sequences, making it difficult to assign their function. To this end, we present a multi-omic epigenetic atlas of the adult human brain through profiling of the chromatin accessibility landscapes and three-dimensional chromatin interactions of seven brain regions across a cohort of 39 cognitively healthy individuals. Single-cell chromatin accessibility profiling of 70,631 cells from six of these brain regions identifies 24 distinct cell clusters and 359,022 cell type-specific regulatory elements, capturing the regulatory diversity of the adult brain. We develop a machine learning classifier to integrate this multi-omic framework and predict dozens of functional single nucleotide polymorphisms (SNPs), nominating gene and cellular targets for previously orphaned GWAS loci. These predictions both inform well-studied disease-relevant genes, such as BIN1 in microglia for Alzheimers disease (AD) and reveal novel gene-disease associations, such as STAB1 in microglia and MAL in oligodendrocytes for Parkinsons disease (PD). Moreover, we dissect the complex inverted haplotype of the MAPT (encoding tau) PD risk locus, identifying ectopic enhancer-gene contacts in neurons that increase MAPT expression and may mediate this disease association. This work greatly expands our understanding of inherited variation in AD and PD and provides a roadmap for the epigenomic dissection of noncoding regulatory variation in disease.

genomics↗

bioRxiv2019

Learning cis-regulatory principles of ADAR-based RNA editing from CRISPR-mediated mutagenesis

Adenosine-to-inosine (A-to-I) RNA editing catalyzed by ADAR enzymes occurs in double-stranded RNAs (dsRNAs). How the RNA sequence and structure (i.e., the cis-regulation) determine the editing efficiency and specificity is poorly understood, despite a compelling need towards functional understanding of known editing events and transcriptome engineering of desired adenosines. We developed a CRISPR/Cas9-mediated saturation mutagenesis approach to generate comprehensive libraries of point mutations near an editing site and its editing complementary sequence (ECS) at the endogenous genomic locus. We used machine learning to integrate diverse RNA sequence features and computationally predicted structures to model editing levels measured by deep sequencing and identified cis-regulatory features of RNA editing. As proof-of-concept, we applied this integrative approach to three editing substrates. Our models explained over 70% of variation in editing levels. The models indicate that RNA sequence and structure features synergistically determine the editing levels. Our integrative approach can be broadly applied to any editing site towards the goal of deciphering the RNA editing code. It also provides guidance for designing and screening of antisense RNA sequences that form dsRNA duplex with the target transcript for ADAR-mediated transcriptome engineering.

molecular biology↗

bioRxiv2019

A genome-wide almanac of co-essential modules assigns function to uncharacterized genes

A central remaining question in the post-genomic era is how genes interact to form biological pathways. Measurements of gene dependency across hundreds of cell lines have been used to cluster genes into co-essential pathways, but this approach has been limited by ubiquitous false positives. Here, we develop a statistical method that enables robust identification of gene co-essentiality and yields a genome-wide set of functional modules. This almanac recapitulates diverse pathways and protein complexes and predicts the functions of 102 uncharacterized genes. Validating top predictions, we show that TMEM189 encodes plasmanylethanolamine desaturase, the long-sought key enzyme for plasmalogen synthesis. We also show that C15orf57 binds the AP2 complex, localizes to clathrin-coated pits, and enables efficient transferrin uptake. Finally, we provide an interactive web tool for the community to explore the results (coessentiality.net). Our results establish co-essentiality profiling as a powerful resource for biological pathway identification and discovery of novel gene functions.

genomics↗

bioRxiv2019

Deep learning at base-resolution reveals motif syntax of the cis-regulatory code

The arrangement of transcription factor (TF) binding motifs (syntax) is an important part of the cis-regulatory code, yet remains elusive. We introduce a deep learning model, BPNet, that uses DNA sequence to predict base-resolution ChIP-nexus binding profiles of pluripotency TFs. We develop interpretation tools to learn predictive motif representations and identify soft syntax rules for cooperative TF binding interactions. Strikingly, Nanog preferentially binds with helical periodicity, and TFs often cooperate in a directional manner, which we validate using CRISPR-induced point mutations. Our model represents a powerful general approach to uncover the motifs and syntax of cis-regulatory sequences in genomics data. HighlightsO_LIThe neural network BPNet accurately predicts TF binding data at base-resolution. C_LIO_LIModel interpretation discovers TF motifs and TF interactions dependent on soft syntax. C_LIO_LIMotifs for Nanog and partners are preferentially spaced at [~]10.5 bp periodicity. C_LIO_LIDirectional cooperativity is validated: Sox2 enhances Nanog binding, but not vice versa. C_LI

genomics↗

bioRxiv2019

Targeted optimization of regulatory DNA sequences with neural editing architectures

Targeted optimizing of existing DNA sequences for useful properties, has the potential to enable several synthetic biology applications from modifying DNA to treat genetic disorders to designing regulatory elements to fine tune context-specific gene expression. Current approaches for targeted genome editing are largely based on prior biological knowledge or ad-hoc rules. Few if any machine learning approaches exist for targeted optimization of regulatory DNA sequences.\n\nHere, we propose a novel generative neural network architecture for targeted DNA sequence editing - the EDA architecture - consisting of an encoder, decoder, and analyzer. We showcase the use of EDA to optimize regulatory DNA sequences to bind to the transcription factor SPI1. Compared to other state-of-the-art approaches such as a textual variational autoencoder and rule-based editing, EDA significantly improves predicted binding of SPI1 of genomic sequences with the minimal set of edits. We also use EDA to design regulatory elements with optimized grammars of CREB1 binding sites that can tune reporter expression levels as measured by massively parallel reporter assays (MPRA). We analyze the properties of the binding sites in the edited sequences and find patterns that are consistent with previously reported grammatical rules which tie gene expression to CRE binding site density, spacing and affinity.

bioinformatics↗

bioRxiv2019

Integrating regulatory DNA sequence and gene expression to predict genome-wide chromatin accessibility across cellular contexts

MotivationGenome-wide profiles of chromatin accessibility and gene expression in diverse cellular contexts are critical to decipher the dynamics of transcriptional regulation. Recently, convolutional neural networks (CNNs) have been used to learn predictive cis-regulatory DNA sequence models of context-specific chromatin accessibility landscapes. However, these context-specific regulatory sequence models cannot generalize predictions across cell types.\n\nResultsWe introduce multi-modal, residual neural network architectures that integrate cis-regulatory sequence and context-specific expression of trans-regulators to predict genome-wide chromatin accessibility profiles across cellular contexts. We show that the average accessibility of a genomic region across training contexts can be a surprisingly powerful predictor. We leverage this feature and employ novel strategies for training models to enhance genome-wide prediction of shared and context-specific chromatin accessible sites across cell types. We interpret the models to reveal insights into cis and trans regulation of chromatin dynamics across 123 diverse cellular contexts.\n\nAvailabilityThe code is available at https://github.com/kundajelab/ChromDragoNN\n\nContactakundaje@stanford.edu

genomics↗

bioRxiv2019

Cell cycle dynamics of human pluripotent stem cells primed for differentiation

Understanding the molecular properties of the cell cycle of human pluripotent stem cells (hPSCs) is critical for effectively promoting differentiation. Here, we use the Fluorescence Ubiquitin Cell Cycle Indicator (FUCCI) system adapted into hPSCs and perform RNA-sequencing on cell cycle sorted hPSCs primed and unprimed for differentiation. Gene expression patterns of signaling factors and developmental regulators change in a cell cycle-specific manner in cells primed for differentiation without altering genes associated with pluripotency. Furthermore, we identify an important role for PI3K signaling in regulating the early transitory states of hPSCs towards differentiation.

genetics↗

bioRxiv2019

Identification and mitigation of pervasive off-target activity in CRISPR-Cas9 screens for essential non-coding elements

Pooled CRISPR-Cas9 screens have recently emerged as a powerful method for functionally characterizing regulatory elements in the non-coding genome, but off-target effects in these experiments have not been systematically evaluated. Here, we conducted a genome-scale screen for essential CTCF loop anchors in the K562 leukemia cell line. Surprisingly, the primary drivers of signal in this screen were single guide RNAs (sgRNAs) with low specificity scores. After removing these guides, we found that there were no CTCF loop anchors critical for cell growth. We also observed this effect in an independent screen fine-mapping the core motifs in enhancers of the GATA1 gene. We then conducted screens in parallel with CRISPRi and CRISPRa, which do not induce DNA damage, and found that an unexpected and distinct set of off-targets also caused strong confounding growth effects with these epigenome-editing platforms. Promisingly, strict filtering of CRISPRi libraries using GuideScan specificity scores removed these confounded sgRNAs and allowed for the identification of essential enhancers, which we validated extensively. Together, our results show off-target activity can severely limit identification of essential functional motifs by active Cas9, while strictly filtered CRISPRi screens can be reliably used for assaying larger regulatory elements.

genomics↗