Rane Lab | Computational Immunology at Columbia University

Welcome to the Rane Lab at Columbia University. We build mechanistic frameworks to understand B and T cell dynamics across developmental, homeostatic, and response timelines.

By integrating physics/ecology-driven principles, deep generative inference (VAEs, Transformers), and experimental multiomics/imaging, we model continuous cell-state transitions and spatial interactions.

New Preprint • bioRxiv (2026)

Quorum sensing underlies the establishment and maintenance of marginal zone B cells

A. Singh, M. Verheijn, T. Hogan, A.J. Yates, B. Seddon*, S. Rane*

We show how density-dependent cellular feedback mechanisms govern marginal zone B cell population dynamics throughout development and homeostasis.

Read on bioRxiv →

Preprint • arXiv (2025)

Spatiodynamic inference using vision-based generative modeling

J.W. Park, K. Zhao, S. Rane

Introducing a computer-vision framework leveraging deep generative modeling to solve inverse problems in complex biological spatiotemporal dynamics.

Read on arXiv →

Lab Images

Opportunities

We are seeking motivated graduate research assistants interested in multidisciplinary projects at the intersection of immunology, machine learning, and mechanistic modeling!

Contact Us

Research Focus

Generative Inference

Deep Generative Inference & Trajectory Optimization
To move beyond simple discrete-compartment models, we develop continuous representations of cell states using time-conditioned Variational Autoencoders (VAEs) and Itô stochastic differential equations (SDEs). Furthermore, our framework—viaABC (Variational Inference-assisted Approximate Bayesian Computation)—distills high-dimensional spatiotemporal imaging and single-cell timeseries to fit and select mechanistic drift-diffusion models.

Lymphocyte Dynamics

Lymphocyte Development & Homeostasis
Using Cre-lox lineage tracing, fate-reporter mouse systems, and mathematical modeling, we quantify T and B cell regulation:

Naive T Cell Lifespan: Post-thymic cell age dictates life expectancy, challenging classical feedback regulation models.
Treg Lineage Mapping: Uncovering a linear pathway of regulatory T cell (Treg) development from thymic origin to memory and tumor-infiltrating subsets.
B Cell Homeostasis & Activation: Identifying extrinsic driver rules governing aging immunoglobulin repertoires and activation-triggered marginal zone B cell fates.

Enhancing Antibody-mediated Tumor Immunity

Clone-Constrained Transcriptional Trajectories
Using advanced computational methods, we analyze the transcriptional dynamics of B and T cells in response in tumors. We combine language transformers, VAEs, and dual-labeling reporter mouse models to map B cell activation histories and clonal divergence in tumors. Our models integrate paired single-cell RNA and V(D)J sequencing data to infer differentiation trajectories constrained by clonal trees and temporal activation windows. The ultimate goal is to identify molecular programs and gene regulatory networks that enhance antibody-mediated tumor immunity.

Principal Investigator

Sanket Rane, PhD is an immunologist turned mathematical modeler, trained in deterministic and probabilistic mathematical modeling, machine learning, Bayesian inference, and cell biology. Sanket earned his PhD in immunology at the National Institute of Immunology in India, investigating molecular and cellular factors governing T cell kinetics during healthy aging. He pivoted to using computational approaches to study lymphocyte ecology during his postdoctoral years at the University of Glasgow and Columbia University. His overarching interest lies in understanding the origins of variability in immune responses, particularly the mechanisms that shape antibody-mediated responses to vaccines, infections, and cancer.

Current Members

Jessica Kinglsey, PhD

Postdoctoral Research Scientist

Builds mathematical and computational models of biological systems across oncology, immunology, and infectious diseases. Current focus is on analyzing single-cell RNA and V(D)J data to better understand B cell immunology and clonal trajectories.

Marek Kampa

Graduate Student (Data Science)

Explores methodologies for projecting high-dimensional B cell dynamics into informative manifolds by integrating deep learning models (VAEs) with conditional modeling, and dynamically guidedlatent space regularization to capture contextual dependencies in biological systems more effectively.

Kangyu Zhao

Graduate Student (Statistics)

Works on enhancing deep learning algorithms that leverage language models and variational inference to model sequential high-dimensional data. His current focus is on developing spatially resolved models to capture CAR T-cell mediated movement and killing of tumor cells in 3D environments.

Alumni

Apoorva Singh Postdoctoral Researcher - Developed mechanistic models for B cell development and homeostasis.
Christine Sharabun Graduate Student (Data Science) - Developed bioinformatics pipelines for single-cell RNA and immune repertoire sequencing datasets.
Jun Won Park Graduate Student (Data Science) - Developed deep learning framework for ill-posed Bayesian inverse problems (viaABC).
Max Sebrechts (Columbia University, Data Science) — VAE latent space representations for B cell differentiation.
Puqi Song (Columbia University, Statistics) — Quantitative visualization and dynamics of TCR repertoires.
Jiayi Dong (Columbia University, Data Science) — Transformer-VAE architectures for spatial biological datasets.

Undergraduate Students

Allison Pascual (Stony Brook University, Biology) — Mathematical modeling of age-associated Fas expression in T cells.
Julia Rehring (University of Michigan) — Bayesian validation of B cell differentiation plasticity.
Danielle (Columbia University, Data Science) — Deep generative modeling of B cell differentiation trajectories.

Preprints

2025–2026

Singh A, Verheijn M, Hogan T, Yates AJ, Seddon B*, Rane S*^#. Quorum sensing underlies the establishment and maintenance of marginal zone B cells. bioRxiv (2026). DOI: 10.1101/2026.01.01.456789. Under review in PNAS.

Park JW, Zhao K, Rane S^#. Spatiodynamic inference using vision-based generative modeling. arXiv (2025). DOI: 10.48550/arXiv.2507.22256.

Featured Publications

Journals

Pathak S*, Hogan T*, Rane S*, Huang Y, Pearson C, Sinclair C, Barry S, Carnevalli L, Yates AJ, Seddon B. A linear ontogeny accounts for the development of naive, memory, and tumor-infiltrating regulatory T cells in mice. Science Immunology (2025). DOI: 10.1126/sciimmunol.adu7341.

Babushku T, Lechner M, Ehrenberg S, Rambold U, Schmidt-Supprian M, Yates AJ, Rane S*, Zimber-Strobl U*, Strobl LJ*. Notch2 controls developmental fate choices between germinal center and marginal zone B cells upon immunization. Nature Communications (2024). DOI: 10.1038/s41467-024-46024-1.

Rane S, Hogan T, Lee E, Seddon B, Yates AJ. Towards a unified model of naive T cell dynamics across the lifespan. eLife (2022). DOI: 10.7554/eLife.78168.

Verheijen M*, Rane S*, Pearson C, Yates AJ, Seddon B. Fate Mapping Quantifies the Dynamics of B Cell Development and Activation throughout Life. Cell Reports (2020). DOI: 10.1016/j.celrep.2020.108376.

Mold JE, Reu P, Olin A, Bernard S, Michaelsson J, Rane S, Yates A, Khosravi A, Salehpour M, Possnert G, Brodin P, Frisen J. Cell generation dynamics underlying naive T cell homeostasis in adult humans. PLOS Biology (2019). DOI: 10.1371/journal.pbio.3000383.

Hayes AJ, Rane S, Scales HE, Meehan GR, Benson RA, Maroof A, Schroeder J, Tomura M, Gozzard N, Yates AJ, Garside P, Brewer JM. Spatiotemporal Modeling of the Key Migratory Events During the Initiation of Adaptive Immunity. Frontiers in Immunology (2019). DOI: 10.3389/fimmu.2019.00598.

Rane S, Hogan T, Seddon B, Yates AJ. Age is not just a number: Naive T cells increase their ability to persist in the circulation over time. PLOS Biology (2018). DOI: 10.1371/journal.pbio.2003949.

Rane S, Das R, Ranganathan V, Prabhu S, Das A, Mattoo H, Durdik JM, George A, Rath S, Bal V. Peripheral residence of naive CD4 T cells induces MHC class II-dependent alterations in phenotype and function. BMC Biology (2014). DOI: 10.1186/s12915-014-0106-0.

^# Corresponding Author; * Equal Contributions

Book chapters

Seddon B, Rane S, Yates AJ. Modelling Naive T Cell Homeostasis. In:Mathematical, Computational, and Experimental T Cell Immunology. Springer (2021).