I am a Senior ML Scientist at Genentech. I am interested in building models to predict cellular behaviour. I completed my PhD in Computer Science at Stanford University, working on single-cell epigenomics and deep learning models for chromatin accessibility in the lab of Anshul Kundaje.

Previously, I have interned at Apple in the Siri team, and developed open-source implementations of interesting machine learning systems such as Alpha Zero and sequence GANs. I received my Bachelors in Electrical Engineering from IIT Delhi.

Photo of Surag Nair

Updates

  • 30 November 2023 I defended my PhD! I'm starting as a Senior ML Scientist at Genentech. [thesis]
  • 21 October 2023 Our preprint on the single-cell dynamics of fibroblast reprogramming to iPSCs is out! [preprint] [tweetorial] [data] [browser] [talk]
  • 24 July 2023 Talk at ISMB 2023 about the sensitivity of regulatory DNA models to spurious features [talk]
  • 14 October 2022 dynseq is published in Nature Genetics! [paper]
  • 8 June 2022 dynseq track preprint out, available at UCSC, WashU and HiGlass/Resgen browsers! [preprint]
  • 8 March 2022 fastISM is published in Bioinformatics! [paper] [code] [docs] [slides] [talk]
  • 19 September 2020 Developed fastISM, 10x faster in-silico saturation mutagenesis for convolutional neural networks. Accepted to MLCB 2020 for oral presentation (15% acceptance rate) [code] [docs] [preprint] [slides] [talk]
  • 29 April 2019 Our paper on a predicting chromatin accessibility across cellular contexts was accepted for an oral presentation in ISMB 2019 [slides] [talk] [code]
  • 29 December 2017 A Simple Alpha(Go) Zero Tutorial