Tarak Nath Nandi preview image

Tarak Nath Nandi

Assistant Computational Scientist

Biography

Tarak Nandi is an Assistant Computational Scientist in the Data Sciences and Learning division within the Computing, Environment, and Life Sciences (CELS) directorate at Argonne, and scientist at large at the University of Chicago Consortium for Advanced Science and Engineering (UChicago CASE). He works at the intersection of computational genomics, cancer biology, high performance computing (HPC) and artificial intelligence (AI).

His work primarily focuses on:

  • In silico prediction of gene and small‑molecule perturbations for drug repurposing: Identifying actionable targets in blood malignancies that may progress to AML/CVD, and repurposing FDA approved drugs to reverse diseased phenotypes by understanding their MOA and from AI predicted transcriptional perturbations.
     
  • De novo protein generation from natural language prompts: Pretraining, finetuning, and RL based alignment of a discrete diffusion model for generating novel proteins with desired functional properties
     
  • Multimodal learning from imaging and transcriptomic data to predict lung cancer survival time
     
  • Identification of spatial niches, cell-to-cell communication (neuron-tumor, tumor-immune) within the Glioblastoma TME using spatial transcriptomic and proteomic data, and biomarkers associated with recurrence time.
     
  • Identification of metabolic biomarkers for lung cancer prognosis and treatment outcome, and understanding the crosstalk between signaling pathways and lung cancer metabolism. 
     
  • Foundation model development: Developing general purpose foundation models for scRNASeq and MRI data, and for gut microbiome metagenome for studying association with CVD and other chronic conditions through inflammatory pathways.
     
  • Developing agentic framework for streamlining and automating hypothesis validation pipelines, with one of the focus areas being the study of low-dose radiation effects on human lung epithelial cells
     
  • Structural variant calling: Leveraging short-read and long-read genome sequencing data to detect large genomic alterations that disrupt gene function.
     
  • Genome-wide association studies (GWAS): Accelerating GWAS to uncover statistical correlations between genetic variants and traits across diverse ancestries for advancing precision medicine.

He is a member of the NIH/NCI Virtual Human Global Initiative, the AMIA complex variant working group, the NIH/NCI computational cancer community User group, and the TOPMed structural variant calling working group.

His past experience includes fundamental and applied research in unsteady aerodynamics, turbulent/transitional flows, atmospheric boundary layer flows using Computational Fluid Dynamics (CFD). He is also experienced in developing deep learning models for gas-solid multiphase flows and reacting flows, and generation of synthetic turbulence signals.

Research Interests

  1. Computational genomics (structural variant calling, population genetics, single cell genomics)
  2. Cancer biology
  3. Scientific computing
  4. Machine learning for scientific applications
    • biomedicine
    • fluid mechanics (multiphase flows, reacting flows)
    • material informatics
    • physics informed neural networks
    • foundation model pretraining, finetuning, and reinforcement learning based aligning
  5. Statistical methods and uncertainty quantification
  6. Unsteady aerodynamics
  7. Turbulent/transitional flow physics and modeling

Selected publications and presentations:

  • (Journal paper) Nandi T, Herrig A and Brasseur J, ​“Non-steady wind turbine response to daytime atmospheric turbulence”, Philosophical Transactions of the Royal Society A (2017)
  • (Journal paper) Simiu E, Potra F and Nandi T, ​“Determining longitudinal integral turbulence scales in the near-neutral atmospheric surface layer”, Boundary Layer Meteorology (2018)
  • (Journal paper) Nandi T and Yeo D, ​“Estimation of integral length scales across the neutral atmospheric boundary layer depth: A Large Eddy Simulation study”, J of Wind Engineering and Industrial Aerodynamics (2021)
  • (Oral presentation) Jordan T, Woo M, Nandi T and Essendelft D, ​“MFiX-AI: A complete cognitive CFD simulation platform for HPC and AI built on TensorFlow”, NVIDIA GPU Technology Conference (April 2021)
  • (Journal paper) Woo M, Jordan T, Nandi T, Dietiker J, Guenther C and Essendelft D, ​“Development of an equation-based parallelization method for multiphase particle-in-cell simulations”, Engineering with Computers (2022)
  • (Oral presentation) Nandi T, Hennigh O, Nabian M, Liu Y, Woo M, Jordan T, Shahnam C, Guenther C and Essendelft D, ​“Developing digital twins for energy applications using Modulus”, NVIDIA GPU Technology Conference (March 2022)
  • (Journal paper) Nandi T, Pintar A and Simiu E, ​“Influence of surface roughness uncertainties on design of structures with open and suburban exposures”, ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A (2022)
  • (Journal paper) Saidi W, Nandi T and Yang T, ​“Designing multinary noble metal-free catalyst for hydrogen evolution reaction”, Electrochemical Science Advances (2022)
  • (Journal paper) Nations S, Nandi T, Ramazani A, Wang S and Duan Y, ​“Metal hydride composition-derived parameters as machine learning features for material design and H2 storage”, Journal of Energy Storage (2023)
  • (Journal paper) Nandi T and Yeo D, ​“A solution verification study for URANS simulations of flow over a 5: 1 rectangular cylinder using grid convergence index and least squares procedures”, ASME Journal of Verification, Validation and Uncertainty Quantification (2023)
  • (Poster presentation) Nandi, T, Popic V, Pankratz N, Rodriguez A, Vats P, Huffman J, Tsao P and Madduri R, ​“Towards an accurate and efficient structural variant calling workflow on leadership scale HPC systems.” 2023 MVP Science Meeting (2023)
  • (Poster presentation) Nandi, T, Theodoris C, Rodriguez A and Madduri R. ​“Prediction of key metastatic genes in head and neck squamous cell carcinoma using a deep learning based context-aware foundation model for network biology.” 9th Computational Approaches for Cancer Workshop held in conjunction with the Supercomputing 2023 conference (2023)
  • (Journal paper) Ramazani A, Duell B, Popczun E, Natesakhawat S, Nandi T, Lekse J, Duan Y, ​“High Throughput Ab-Initio Calculations and Machine Learning Modeling to Design and Discover SrFeO3-δ -based Perovskite Materials for Chemical Looping Applications”, Cell Reports Physical Science (2024)
  • (Journal paper) Nandi T, Chong L, Park J, Saidi W, Chorpening B, Bayham S and Duan Y, ​“A Machine Learning Approach for Determining Temperature-Dependent Band Gap of Metal Oxides Utilizing Allen-Heine-Cardona Theory and O’Donnell Model Parameterization”, AIP Advances (2024)
  • (Journal paper) Carrillo-Perez F, Pizurica M, Zheng Yuanning, Nandi T, Madduri R, Shen J and Gevaert O, ​“RNA-to-image multi-cancer synthesis using cascaded diffusion models”, Nature Biomedical Engineering (2024)
  • (Journal paper) Verma et al., ​“Diversity and scale: genetic architecture of 2,068 traits in the VA Million Veteran Program”, Science (2024)
  • (Oral presentation) Nandi T, Claybon J, Kashyap S, Conery M, Rodriguez A, Li Z, Wu J and Madduri R, ​“An AI Agentic Framework for Understanding Low-Dose Radiation Effects on Human Lung Epithelial Cells”, oral presentation at the Eleventh Computational Approaches for Cancer Workshop, CAFCW25 (2025)
  • (Journal paper) Chen H, Venkatesh M, Ortega J, Mahesh S, Nandi T, Madduri R, Pelka K, and Theodoris C ​“Quantized multi-task learning for context-specific representations of gene network dynamics”, Nature Computational Science (2026)
  • (Oral presentation) Louw J (presenter), Nandi T, Calabrese E, McLendon R, Sun Y, Hickey J, Jepson J, Corcoran A, Zhang K, Madduri R and Khasraw M, ​“A Multimodal AI framework integrating spatial omics and radiomics for recurrence prediction in glioblastoma”, oral presentation at American Association for Cancer Research Annual Meeting, AACR 2026, San Diego, CA (2026)
  • (Journal paper) Bareja R, Carrillo-Perze F, Zheng Y, Pizurica M, Nandi T, Tian L, Shen J, Madduri R and Gevaert O, ​“A benchmark study of vision and pathology foundation models for computational pathology”, Nature Communications (2026)
  • (preprint) Pershad Y*, Nandi T*, Van Amburg J, Parker A, Ostrowski L, Giannini H, Ong D, Heimlich J, Obeng E, Ericson K, Agarwal A, Madduri R, Bick A, ​“Closing the loop: Teaching single-cell foundation models to learn from perturbations”, bioArxiv preprint (under review in Cell Genomics)

Education

  • PhD in Mechanical Engineering from The Pennsylvania State University (2012-2016)
  • PhD coursework in Aerospace Engineering at Georgia Institute of Technology (transferred to Penn State) (2010-11)
  • MTech in Mechanical Engineering from Indian Institute of Technology Kanpur (2008-2010)
  • BTech in Mechanical Engineering from National Institute of Technology Durgapur (2004-2008)