Ravi Madduri
Has grant
Senior Scientist
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Recent Grants
Grant: Close
Collaborative Research: SI2-SSI: Swift/E: Integrating Parallel Scripted Workflow into the Scientific Software Ecosystem
Open Date: 2016-10-01
Close Date: 2022-09-30
Grant: Close
Hardening Globus Genomics to make large-scale genomics analysis available for everybody
Open Date: 2016-09-28
Close Date: 2020-07-31
Articles (19)
Year:
2026
Year:
2026
Year:
2025
The application of transfer learning models to large scale single-cell datasets has enabled the development of single-cell foundation models (scFMs) that can predict cellular responses to perturbations in silico. Although these predictions can be experimentally tested, current scFMs are unable to “close the loop” and learn from these experiments to create better predictions. Here, we introduce a “closed-loop” framework that extends the scFM by incorporating perturbation data during model fine-tuning. Our closed-loop model improves prediction accuracy, increasing positive predictive value in the setting of T-cell activation three-fold. We applied this model to RUNX1-familial platelet disorder, a rare pediatric blood disorder and identified two therapeutic targets (mTOR and CD74-MIF signaling axis) and two novel pathways (protein kinase C and phosphoinositide 3-kinase). This work establishes that iterative incorporation of experimental data to foundation models enhances biological predictions, representing a crucial step toward realizing the promise of "virtual cell" models for biomedical discovery.
Year:
2025
To advance precision medicine in pathology, robust AI-driven foundation models are increasingly needed to uncover complex patterns in large-scale pathology datasets, enabling more accurate disease detection, classification, and prognostic insights. However, despite substantial progress in deep learning and computer vision, the comparative performance and generalizability of these pathology foundation models across diverse histopathological datasets and tasks remain largely unexamined. In this study, we conduct a comprehensive benchmarking of 31 AI foundation models for computational pathology, including general vision models (VM), general vision-language models (VLM), pathology-specific vision models (Path-VM), and pathology-specific vision-language models (Path-VLM), evaluated over 41 tasks sourced from TCGA, CPTAC, external benchmarking datasets, and out-of-domain datasets. Our study demonstrates that Virchow2, a pathology foundation model, delivered the highest performance across TCGA, CPTAC, and external tasks, highlighting its effectiveness in diverse histopathological evaluations. We also show that Path-VM outperformed both Path-VLM and VM, securing top rankings across tasks despite lacking a statistically significant edge over vision models. Our findings reveal that model size and data size did not consistently correlate with improved performance in pathology foundation models, challenging assumptions about scaling in histopathological applications. Lastly, our study demonstrates that a fusion model, integrating top-performing foundation models, achieved superior generalization across external tasks and diverse tissues in histopathological analysis. These findings emphasize the need for further research to understand the underlying factors influencing model performance and to develop strategies that enhance the generalizability and robustness of pathology-specific vision foundation models across different tissue types and datasets. PathBench : https://pathbench.stanford.edu/
Year:
2025
Large-scale genetic association studies have identified thousands of trait-associated risk loci, establishing the polygenic basis for common complex traits and diseases. Although prior studies suggest that many trait-associated loci are pleiotropic, the extent to which this pleiotropy reflects shared causal variants or confounding by linkage disequilibrium remains poorly characterized. To define a set of candidate loci with potentially pleiotropic associations, we performed genome-wide association study (GWAS) meta-analyses of up to 1,167 clinically relevant traits and diseases across 1,789,365 diverse individuals genetically similar to Admixed American (AMR, N Max = 60,756), African (AFR, N Max = 128,361), East Asian (EAS, N Max = 307,465), European (EUR, N Max = 1,283,907), and South Asian (SAS, N Max = 8,876) reference populations from the VA Million Veteran Program (MVP), UK Biobank (UKB), FinnGen, Biobank Japan (BBJ), Tohoku Medical Megabank (ToMMo), and Korean Genome and Epidemiology Study (KoGES). We identified 27,193 genome-wide significant locus-trait pairs (1MB region with P GWAMA 1 trait. Multi-trait colocalization identified 1,902 (72%) loci with high-confidence (posterior probability > 0.9) evidence of a shared causal variant across two or more traits. Variants in pleiotropic loci were significantly enriched for a broad spectrum of functional annotations compared to non-pleiotropic counterparts. Polygenic scores (PGS) developed from these data generally improved prediction compared to existing PGS, and were broadly associated with both primary and pleiotropic phenotypes. These results provide a contemporary map of genetic pleiotropy across the spectrum of human traits/diseases and diverse genetic backgrounds.
Year:
2025
Year:
2024
Year:
2024
Year:
2024
While often represented as static entities, gene networks are highly context-dependent. Here, we developed a multi-task learning strategy to yield context-specific representations of gene network dynamics. We assembled a corpus comprising ~103 million human single-cell transcriptomes from a broad range of tissues and diseases and performed a two stage pretraining, first with non-malignant cells to generate a foundational model and then with continual learning on cancer cells to tune the model to the cancer domain. We performed multi-task learning with the foundational model to learn context-specific representations of a broad range of cell types, tissues, developmental stages, and diseases. We then leveraged the cancer-tuned model to jointly learn cell states and predict tumor-restricting factors within the colorectal tumor microenvironment. Model quantization allowed resource-efficient fine-tuning and inference while preserving biological knowledge. Overall, multi-task learning enables context-specific disease modeling that can yield contextual predictions of candidate therapeutic targets for human disease.
Year:
2024
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Collaborators (38)
Joseph Lachance
Associate Professor
Georgia Institute of Technology
Anurag Verma
Assistant Professor
University of Pennsylvania
Robert Klein
Associate Professor
Icahn School of Medicine at Mount Sinai
Katherine Liao
Associate Professor
Harvard Medical School
Florence Menegaux
Research Director / MD - PhD
INSERM
Angelica Walker
Assistant Professor
George Washington University
Lynda Brinson
Professor
Duke University
Sarah Abramowitz
Stanford University
Rebecca Graff
Associate Professor
University of California, San Francisco
Michael Garvin
Staff Scientist
Oak Ridge National Laboratory
Hilary Coon
Professor
University of Utah
Dokyoon Kim
Assistant Professor
University of Pennsylvania
Scott M Damrauer
Assistant Professor
University of Pennsylvania Perelman School of Medicine
Mitchell Machiela
Senior Investigator
National Cancer Institute
Satoshi Koyama
Associate Member
Broad Institute
Kibaek Kim
Senior Scientist at-Large
University of Chicago
Brandon L Pierce
Assistant Professor
University of Chicago
Pradeep Natarajan
Associate Professor
Harvard Medical School
Jean Luca Bez
Career-Track Research Scientist
Lawrence Berkeley National Laboratory
Geoffrey Fox
Professor
University of Virginia
Anna Starnawska
Aarhus University
Karin Pelka
Gladstone Institutes
Eric Monson
Assistant Professor
University of Utah
Benjamin F. Voight
Professor
University of Pennsylvania
Esther Obeng
Associate Professor
Emory University
Olivier Gevaert
Assistant Professor
Stanford University
Michael Levin
Cardiologist
Corporal Michael J. Crescenz VA Medical Center
Yuanning Zheng
Assistant Professor
University of Houston
Volodymyr Kindratenko
University of Illinois at Urbana-Champaign
Wenlong Chen
Senior Researcher
University of Witwatersrand
Mark Neubauer
Professor
University of Illinois at Urbana-Champaign
Alexander Bick
Assistant Professor
Vanderbilt University
Jonathan Coleman
Senior Lecturer in Statistical Genetics
King’s College London
Eliu Huerta
CASE (Consortium for Advanced Science and Engineering) Senior Scientist
University of Chicago
Luc Multigner
INSERM
Hyejung Won
Associate Professor
University of North Carolina at Chapel Hill
Marcus Sokolowski
Healthcare Provision, Stockholm County
Jeanne Shen
Associate Professor
Stanford University
Make your next move.
Prepare your application
Review drafts before you send them.