Vikram Agarwal

This paper uses genomic foundation models to assess how somatic mutations disrupt cancer biology and clinical outcomes across multiple cancer types.

The paper predicts translation efficiency (TE) of mRNA in mammalian cells using a deep learning model trained on ribosomal profiling data across multiple cell types.

The paper analyzes ribosome profiling data to show translation efficiency covariation (TEC) is conserved across human and mouse cells, revealing gene functions not evident from RNA or protein co-expression.

GAME provides a standardized framework for benchmarking genomics models across tasks, enabling consistent evaluation and community contributions.

MPRAbase is a curated database of MPRA experiments that provides access to 17.7 million regulatory elements tested across 35 cell types and 4 organisms.

The paper identifies functional cis-regulatory elements in three human cell types using lentiMPRAs and develops predictive models for CRE function.

This paper introduces Borzoi, a model that predicts RNA-seq coverage from DNA sequence, enabling variant effect scoring across multiple regulatory layers and outperforming existing methods on QTLs.

CodonBERT is a large language model for mRNA design and optimization that outperforms previous methods on a flu vaccine dataset.

The paper compiles mRNA decay rate datasets from humans and mice, identifies biases, corrects them, and trains a deep learning model to predict half-life from sequence with improved accuracy.

This paper introduces ENGRAM, a method to genomically record enhancer and signal transduction activity in mammalian cells, demonstrating its ability to track multiple transcriptional reporters with high fidelity.

Enformer, a sequence-based transformer integrating elements up to 100 kb away, lifts held-out gene expression prediction (correlation 0.81 to 0.85 versus Basenji2) and variant effect prediction.

The paper quantifies alternative polyadenylation (APA) across 2 million single cells from developing mouse embryos, revealing dynamic changes in 3' UTR length and identifying potential regulators in neuronal and hematopoietic lineages.

The paper shows that MPRA data exhibit strand asymmetry favoring naturally transcribed strands, suggesting biological functions driving this effect across gene bodies and Alu elements.

This study evaluates how experimental design, orientation, and sequence context affect MPRA results, finding significant differences in regulatory activity measurements.

The paper develops Xpresso, a deep convolutional neural network that predicts mRNA abundance from genomic sequence, achieving 59% and 71% variance explained in human and mouse, respectively.

The paper functionally tests osteoarthritis-associated variants to identify regulatory activity, finding six SNPs with differential effects, including rs4730222 increasing HBP1 isoform expression.

The paper profiles CRISPR/Cas9-induced mutational outcomes and develops a predictive model based on sequence context to guide DSB repair outcomes.

The paper develops a quantitative model to predict microRNA targeting efficacy in Drosophila, outperforming existing models and updating TargetScanFly.

The paper characterizes the evolutionary-functional trajectories of liver enhancers in primates by analyzing sequence mutations and their impact on regulatory activity.