Rohit Singh

Rohit Singh Lab: {Publications}

Below, an asterisk (*) by the name of the authors indicates that they were co-first (i.e., equal contributions). A caret (^) indicates that the authors were co-corresponding.

Single-cell Genomics + Causal Inference

Ozbay S, Parekh A, Singh R^. Navigating the manifold of single-cell gene coexpression to discover interpretable gene programs.. Under review, 2023 (paper, via bioRxiv. http://singhlab.net/Sceodesic).
Wu A*, Singh R*^, Walsh CA, Berger B^. An econometric lens resolves cell-state parallax. Under review, 2023 (paper, via bioRxiv. http://cb.csail.mit.edu/cb/gridnet/).
Singh R*^, Wu A*, Mudide A*, Berger B^. Unraveling causal gene regulation from the RNA velocity graph using Velorama. Cell Systems (paper; bioRxiv preprint; http://cb.csail.mit.edu/cb/velorama/). Presented at RECOMB 2023. Invited for consideration by both Cell Systems and Genome Research.

Wu A, Markovich T^, Berger B, Hammerla N, Singh R^. Causally-guided Regularization of Graph Attention Improves Generalizability . TMLR, 2023 (paper, via arXiv). In collaboration with Twitter Research.

Singh R*^, Li J*, Tattikota S*, Liu Y, et al. Optimal transport analysis of single-cell transcriptomics directs hypotheses prioritization and validation. Under review, 2022 (paper, via bioRxiv. http://cb.csail.mit.edu/cb/haystack/).

Singh R^*, Wu A*, Berger B^. Granger causal inference on DAGs identifies genomic loci regulating transcription . Int'l Conf. on Learning Representations (ICLR) 2022 (scored in the top ~1% of submissions) (paper, via arXiv. https://github.com/alexw16/gridnet).

Singh R^, Berger B^. Deciphering the species-level structure of topologically associating domains. Under review, 2021 (paper, via bioRxiv. http://cb.csail.mit.edu/cb/tadmap/).

Singh R*^, Hie B*, Narayan A, Berger B^. Schema: metric learning enables interpretable synthesis of heterogeneous single-cell modalities. Genome Biology, 2021 (paper, via the journal. http://cb.csail.mit.edu/cb/schema/).

Understanding the language of proteins

NaderiAlizadeh N^, Singh R. Aggregating Residue-Level Protein Language Model Embeddings with Optimal Transport. bioRxiv, 2024. (preprint. GitHub).

Singh R*, Im C*, et al. Learning the Language of Antibody Hypervariability. bioRxiv, 2023. (preprint. GitHub).

Sledzieski S*, Devkota K*, Singh R, Cowen L^, Berger B^. TT3D: Leveraging precomputed protein 3D sequence models to predict protein protein interactions. Bioinformatics, 2023. (paper. GitHub).

Singh R*, Sledzieski S*, Cowen L^, Berger B^. Learning the Drug-Target Interaction Lexicon. Under review, 2022 (paper, via bioRxiv. GitHub).

Sledzieski S*, Singh R*, Cowen L^, Berger B^. Contrasting drugs from decoys. NeurIPS MLSB: Workshop on Machine Learning in Structural Biology, 2022 (to appear, bioRxiv ).

Singh R*, Devkota K*, Sledzieski S, Berger B, Cowen L. Topsy-Turvy: integrating a global view into sequence-based PPI prediction. Bioinformatics, 2022. Presented at ISMB 2022. (paper)
Sledzieski S*, Singh R*, Cowen L, Berger B. Adapting Protein Language Models to DTI Prediction. NeurIPS MLSB: Workshop on Machine Learning in Structural Biology, 2021. paper).

Sledzieski S*, Singh R*, Cowen L, Berger B. D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein-protein interactions. Cell Systems, 2021 (paper).

Sledzieski S*, Singh R*, Cowen L, Berger B. Sequence-based prediction of protein-protein interactions: a structure-aware interpretable deep learning model. Proc of Int'l Conf on Research in Comp Mol Bio (RECOMB) 2021 (preprint, via bioRxiv).

Analysis of Protein-Protein Interaction Networks

Park D*, Singh R*, Baym M, Liao CS, Berger B. IsoBase: a database of functionally related proteins across PPI networks.. Nucleic Acids Res. 2011 Jan;39(Database issue):D295-300. (paper, via Pubmed Central. http://isobase.csail.mit.edu).

Singh R*, Park D*, Xu J*, Hosur R, Berger B. Struct2Net: a web service to predict protein-protein interactions using a structure-based approach. Nucleic Acids Res. 2010 Jul;38(Web Server issue):W508-15. (paper, via Pubmed Central. http://struct2net.csail.mit.edu)

Liao CS, Lu K, Baym M, Singh R, Berger B. IsoRankN: spectral methods for global alignment of multiple protein networks. Bioinformatics. 2009 June 15; 25(12): i253�i258. (paper, via Pubmed Central. http://isorank.csail.mit.edu)
A preliminary version of this work was presented at Int'l Conf on Intelligent Systems for Mol Bio (ISMB) 2009.

Kaplow IM*, Singh R*, Friedman A, Bakal C, Perrimon N, Berger B. RNAiCut: automated detection of significant genes from functional genomic screens. Nature Methods. 2009 Jul;6(7):476-7. (paper, via Nature Methods. http://rnaicut.csail.mit.edu)

Singh R, Xu J, Berger B. Global alignment of multiple protein interaction networks with application to functional orthology detection. Proc Natl Acad Sci (PNAS) U S A. 2008 Sep 2;105(35):12763-8 (paper, via PNAS. Citations. http://isorank.csail.mit.edu)

Singh R, Xu J, Berger B. Global alignment of multiple protein interaction networks. Pac Symp Biocomput. (PSB) 2008:303-14 (paper, via PSB. http://isorank.csail.mit.edu)

Singh R, Berger B. Influence Flow: Integrating Pathway-specific RNAi data and Protein Interaction Data. Int'l Symp of Intelligent Sys in Mol Bio (ISMB) 2007, PLoS Track (Extended Abstract)
More detail is available in my PhD thesis. As per the ISMB PLoS Track rules, the extended abstract should be cited as "Personal Communications"

Singh R, Xu J, Berger B. Pairwise Global Alignment of Protein Interaction Networks by Matching Neighborhood Topology. Proc of Int'l Conf on Research in Comp Mol Bio (RECOMB) 2007 (paper. Citations. http://isorank.csail.mit.edu)

Sontag D*, Singh R*, Berger B. Probabilistic modeling of systematic errors in two-hybrid experiments. Pac Symp Biocomput (PSB) 2007. (paper, via PSB. http://groups.csail.mit.edu/cb/probmod2H)

Singh R*, Xu J*, Berger B. Struct2net: integrating structure into protein-protein interaction prediction. Pac Symp Biocomput (PSB) 2006. (paper, via PSB. http://struct2net.csail.mit.edu)

Broader Systems and Structural Biology

Hosur R, Singh R, Berger B. Sparse estimation for structural variability. Algorithms Mol Biol. 2011; 6: 12. (paper, via Pubmed Central)
A preliminary version of this paper was presented at WABI 2010.

Sterner B, Singh R, Berger B. Predicting and annotating catalytic residues: an information theoretic approach. J Comput Biol. 2007 Oct;14(8):1058-7. (paper, via JCB)

Singh R, Palmer N, Gifford D, Berger B, Bar-Joseph Z. Active Learning for Sampling in Time-Series Experiments With Application to Gene Expression Analysis. Proc. of the 22nd International Conference on Machine Learning (ICML), 2005. (paper)

Singh R, Berger B. Chaintweak: sampling from the neighbourhood of a protein conformation. Pac Symp Biocomput (PSB) 2005:52-63. (paper, via PSB. http://theory.csail.mit.edu/chaintweak/)

Singh R, Saha M. Identifying structural motifs in proteins. Pac Symp Biocomput (PSB) 2003:228-39. (paper, via PSB.)

Other Areas

Deshpande H, Nam U, Singh R.Classification of Music Signals In The Visual Domain. Proc of Digital Audio Effects Conf, 2001

Lahoti A, Singh R, Mukerjee A.A Multidimensional Interval Algebra Approach for Identifying Block-Structured Media.Proc of IAPR Int'l Workshop on Graphics Recognition (GREC) 1999. (paper, via LNCS)

Singh R, Lahoti A, Mukerjee A.Document Block Layout Analysis and Document Template Generation. Proc of Document Layout Interpretation and Analysis Workshop (DLIA) 1999 (paper)

PhD and MS Theses:

PhD Thesis: Algorithms for the Analysis of Protein Interaction Networks ( PDF | Defense Talk )

Winner of the George M. Sprowls Award for the Best PhD Theses in Computer Science at MIT

MS Thesis: Identifying Structural Motifs in Proteins ( PDF )

Winner of the Christopher Stephenson Award for the Best M.S. Thesis in Computer Science at Stanford University

Patent(s)

Allen Chefitz and Rohit Singh. Self-administered, Non-invasive, Transcutaneous Viral Detector. USPTO 17/079649. Filed on Oct 26, 2022. Issued on Jan 20, 2022.

Bonnie Berger Leighton and Rohit Singh. Method for identifying network similarity by matching neighborhood topology. USPTO #8000262. Filed on Apr 18, 2008; issued on Aug 16, 2011. The patent covers much of the core technology around the IsoRank algorithm.

XHTML | CSS Web site Design by : WDD