Prateek Anand

UCLA Computer Science Ph.D. Student

prof_pic_color_square_zoomed.jpg

prateek@cs.ucla.edu

(408) 431-1325

Los Angeles, CA

I am a Ph.D. student in Computer Science at UCLA, advised by Sriram Sankararaman. I work on machine learning for genomics, with a focus on deep generative models.

I am currently developing discrete diffusion models for cell-type-specific regulatory DNA design, building generative models of human genetic variation for synthetic data and imputation, and exploring ensemble machine learning approaches for imputing rare variants across diverse ancestries. I also work on scalable statistical methods for understanding human genetic architecture.

news

Oct 06, 2026 Genetic Probabilistic Circuits (GPC) is now published in PLOS Genetics!
Apr 30, 2026 I was selected as one of four students nominated by UCLA for the Google PhD Fellowship!
Mar 02, 2026 Our work on Genetic Probabilistic Circuits (GPC) has been accepted as an Oral Spotlight (Top 10%) at the ICLR 2026 Gen² Workshop. See you in Brazil!
Oct 15, 2025 I will be presenting on deep generative models for genetics at #ASHG2025 in Boston!

selected publications

  1. plos_genetics.jpg
    GPC: An expressive and tractable deep generative model for genetic variation data
    Prateek Anand, Anji Liu, Meihua Dang, Boyang Fu, Xinzhu Wei, Guy Van den Broeck*, and Sriram Sankararaman*
    PLOS Genetics, Oct 2026
  2. nature_genetics.jpg
    A biobank-scale test of marginal epistasis reveals genome-wide signals of polygenic interaction effects
    Boyang Fu*, Ali Pazokitoroudi*, Zhuozheng Shi, Asha Kar, Albert Xue, Aakarsh Anand, Prateek Anand, Zhengtong Liu, Richard Border, Paivi Pajukanta, Noah Zaitlen, and Sriram Sankararaman
    Nature Genetics, 2025
  3. genome_research.jpg
    A scalable adaptive quadratic kernel method for interpretable epistasis analysis in complex traits
    Boyang Fu*, Prateek Anand*, Aakarsh Anand*, Joel Mefford, and Sriram Sankararaman
    Genome Research, 2024
The branching tree learned by GPC, drawn as DNA strands inside a ring of sequence letters
The branching tree learned by GPC, drawn as DNA strands inside a ring of sequence letters

Most genetic models treat DNA as a simple chain. GPC instead learns the branching tree shown here, which connects positions that can be far apart, and builds a deep generative model on top of it.