Biology subjects

Agarwal, V.

Publications and source records attributed to Agarwal, V..

3 recordsLinked to original sources

bioRxiv2018

Predicting mRNA abundance directly from genomic sequence using deep convolutional neural networks

Algorithms that accurately predict gene structure from primary sequence alone were transformative for annotating the human genome. Can we also predict the expression levels of genes based solely on genome sequence? Here we sought to apply deep convolutional neural networks towards this goal. Surprisingly, a model that includes only promoter sequences and features associated with mRNA stability explains 59% and 71% of variation in steady-state mRNA levels in human and mouse, respectively. This model, which we call Xpresso, more than doubles the accuracy of alternative sequence-based models, and isolates rules as predictive as models relying on ChIP-seq data. Xpresso recapitulates genome-wide patterns of transcriptional activity and predicts the influence of enhancers, heterochromatic domains, and microRNAs. Model interpretation reveals that promoter-proximal CpG dinucleotides strongly predict transcriptional activity. Looking forward, we propose the accurate prediction of cell type-specific gene expression based solely on primary sequence as a grand challenge for the field.

genomics↗

bioRxiv2018

Functional Characterization of Enhancer Evolution in the Primate Lineage

BackgroundEnhancers play an important role in morphological evolution and speciation by controlling the spatiotemporal expression of genes. Due to technological limitations, previous efforts to understand the evolution of enhancers in primates have typically studied many enhancers at low resolution, or single enhancers at high resolution. Although comparative genomic studies reveal large-scale turnover of enhancers, a specific understanding of the molecular steps by which mammalian or primate enhancers evolve remains elusive.\n\nResultsWe identified candidate hominoid-specific liver enhancers from H3K27ac ChIP-seq data. After locating orthologs in 11 primates spanning [~]40 million years, we synthesized all orthologs as well as computational reconstructions of 9 ancestral sequences for 348 \"active tiles\" of 233 putative enhancers. We concurrently tested all sequences (20 per tile) for regulatory activity with STARR-seq in HepG2 cells, with the goal of characterizing the evolutionary-functional trajectories of each enhancer. We observe groups of enhancer tiles with coherent trajectories, most of which can be explained by one or two mutational events per tile. We quantify the correlation between the number of mutations along a branch and the magnitude of change in functional activity. Finally, we identify 57 mutations that correlate with functional changes; these are enriched for cytosine deamination events within CpGs, compared to background events.\n\nConclusionsWe characterized the evolutionary-functional trajectories of hundreds of liver enhancers throughout the primate phylogeny. We observe subsets of regulatory sequences that appear to have gained or lost activity at various positions in the primate phylogeny. We use these data to quantify the relationship between sequence and functional divergence, and to identify CpG deamination as a potentially important force in driving changes in enhancer activity during primate evolution.

genomics↗

bioRxiv2017

Predicting microRNA targeting efficacy in Drosophila

Important for understanding the regulatory roles of miRNAs is the ability to predict the mRNA targets most responsive to each miRNA. Here, we acquired datasets needed for the quantitative study of microRNA targeting in Drosophila. Analyses of these data expanded the types of sites known to be effective in flies, expanded the mRNA regions with detectable targeting to include 5' UTRs, and identified features of site context that correlate with targeting efficacy. Updated evolutionary analyses evaluated the probability of conserved targeting for each predicted site and indicated that more than a third of the Drosophila genes are preferentially conserved targets of miRNAs. Based on these results, a quantitative model was developed to predict targeting efficacy in insects. This model performed better than existing models and will drive the next version of TargetScanFly (v7.0; targetscan.org), thereby providing a valuable resource for placing miRNAs into gene-regulatory networks of this important experimental organism.

systems biology↗