Aarthi Venkat

Postdoctoral Fellow in the Schmidt Center @ Broad Institute of MIT and Harvard

Aarthi Venkat is a Postdoctoral Fellow in the Schmidt Center at the Broad Institute of MIT and Harvard. Co-mentored by Drs. Marinka Zitnik and Nir Hacohen, her work advances geometric machine learning foundations toward a deeper understanding of patient biology, focusing on learning molecular and cellular insights. Prior to her postdoctoral work, Aarthi completed her Ph.D. at Yale University with Dr. Smita Krishnaswamy, where she developed new methodology for single-cell analysis, collaborating across diverse disciplines to generate biologically-plausible hypotheses and actionable scientific insights. Her work has appeared in journals such as Cancer Discovery, Nature Computational Science, and Genome Research, as well as in computer science conferences such as ICASSP and LoG.

Selected Projects

AAnet resolves a continuum of spatially-localized cell states to unveil tumor complexity

AAnet is a neural network for nonlinear archetypal analysis of single-cell data.

A Venkat*, S Youlten*, BP San Juan*,…,S Krishnaswamy, CL Chaffer. Cancer Discovery 2025.

A reservoir of stem-like CD8+ T cells in the tumor-draining lymph node preserves the ongoing antitumor immune response

Stem-like CD8+ T cells in dLN maintain stemness and protect antitumor T cells from terminal differentiation in the TME.

KA Connolly, M Kuchroo, A Venkat,…,NS Joshi. Science Immunology 2021.

Beta cells are essential drivers of pancreatic ductal adenocarcinoma development

Disentangling the origin of β cells misexpressing pro-tumorigenic hormones in the context of obesity and other stressors.

C Garcia*, A Venkat*, DC McQuaid*,…,S Krishnaswamy, MD Muzumdar. Nature Communications 2026.

Cross-modal transfer learning for mapping bulk transcriptomes at cellular level

POPPY is a contrastive learning framework that transfers single-cell foundation model representations to bulk transcriptomics via ontology-aware alignment, enabling generalizable patient embeddings and phenotype prediction.

A Venkat, Z Jiang*, D Marbach*, N Hacohen, M Zitnik. ICML 2026 FM4LS Workshop.

Directed Scattering for Knowledge Graph-based Cellular Signaling Analysis

DSAE employs a directed version of a geometric scattering transform for embedding nodes of directed graphs.

A Venkat*, J Chew*, F Cardoso Rodriguez, CJ Tape, M Perlmutter, S Krishnaswamy. ICASSP 2024.

Graph Fourier MMD for Signals on Graphs

GFMMD embeds signals defined on graphs via an optimal witness function.

S Leone, A Tong, G Huguet, A Venkat, G Wolf, S Krishnaswamy. SampTA 2023.

Inferring dynamic regulatory interaction graphs from time series data with perturbations

RiTINI infers time-varying interaction graphs to predict casual behavior of a system. D Bhaskar, S Magruder, E De Brouwer, A Venkat, F Wenkel, G Wolf, S Krishnaswamy. LoG Conference 2023.

Mapping the gene space at single-cell resolution with gene signal pattern analysis

GSPA learns gene-gene relationships from single-cell data by embedding the gene space.

A Venkat, S Leone, S Youlten, E Fagerberg, J Attanasio, NS Joshi, S Krishnaswamy. Nature Computational Science 2024.

Multiscale geometric and topological analyses for characterizing and predicting immune responses from single cell data

This review describes approaches to characterize immune heterogeneity at multiple resolutions.

A Venkat, D Bhaskar, S Krishnaswamy. Cell Trends in Immunology 2023.

PD-1 maintains CD8 T cell tolerance towards cutaneous neoantigens

PD-1 avoids immunopathology by preventing CD8 T cells to attain fully functional state.

M Damo, N Hornick, A Venkat,…,S Krishnaswamy, N Joshi. Nature 2023.

Revealing dynamic temporal trajectories and underlying regulatory networks with Cflows

Cflows learns continuous and cell-specific trajectories from longitudinal single-cell data.

X Sun*, S Gupta*, A Tong*, M Kuchroo*, D Bhaskar, C Liu, A Venkat,…,CL Chaffer, S Krishnaswamy. In Review.

Single-cell analysis reveals transcriptional dynamics in primary parathyroid tissue

The first parathyroid single-cell atlas of primary parathyroid tissue.

A Venkat*, M Carlino*, B Lawton*,…,S Krishnaswamy, D Krause. Genome Research 2024.

Single-cell multi-modal GAN reveals spatial patterns in single-cell data from triple-negative breast cancer

scMMGAN uses adversarial learning to integrate single-cell and spatial modalities.

M Amodio, SE Youlten, A Venkat,…,CL Chaffer, S Krishnaswamy. Cell Patterns 2022.