About

Rahul M. Kohli, M.D., Ph.D., is the Francis C. Wood Associate Professor in Medicine II at the University of Pennsylvania's Perelman School of Medicine. He serves as an Attending Physician in Infectious Diseases at the Hospital of the University of Pennsylvania and is a Penn Scholar in Molecular Medicine within the Department of Medicine. His academic affiliations include Pharmacology, Cell and Molecular Biology, Biochemistry and Molecular Biophysics, and Immunology. Dr. Kohli's research focuses on the chemistry of the genome, specifically on DNA modifying enzymes that add complexity to genomic DNA through processes such as mutation, chemical modification of nucleobases, and epigenetic regulation. His laboratory investigates enzymes involved in enzymatic deamination, oxidation, and methylation of cytosine bases, with particular attention to AID/APOBEC DNA deaminases and TET oxygenases. Additionally, his work explores pathogen pathways that promote evolution and antibiotic resistance, especially the LexA/RecA axis governing the bacterial SOS response. His research employs biochemical characterization, chemical synthesis of enzyme probes, and biological assays across bacteriology, immunology, and virology to understand and manipulate DNA-modifying enzymes and pro-mutagenic pathways, aiming to harness these pathways for biotechnological applications.

Research topics

• Biology
• Biochemistry
• Chemistry
• Computational biology
• Genetics

Selected publications

• Sparse-seq TAE Calculator

  Zenodo (CERN European Organization for Nuclear Research) · 2026-03-26

  otherOpen accessSenior author

  The TAE Calculator for Sparse-Seq, a method of quantifying cytosine modification using low-coverage sequencing https://zhou-lab.github.io/TAE_calculator/

  PublisherDOI

• Unnatural Cytosine Analogs Potentiate a Customizable, Enzymatic Method for Integrated Epigenetic and Four-Base Genetic Sequencing

  Journal of the American Chemical Society · 2026-03-11

  articleOpen accessSenior authorCorresponding

  The interplay of genetic and epigenetic information shapes cell identity, development, and disease. However, standard methods for profiling DNA modifications (e.g., bisulfite sequencing) rely on selective C-to-T conversions, hindering the simultaneous examination of both genetic and epigenetic information. Here, we introduce Integrated Sequencing, which provides high-fidelity mapping of DNA modifications while preserving the native four-base genetic code in single DNA molecules. Integrated-Seq leverages the synthesis of a tethered copy strand with unnatural cytosine analogs that resist enzymatic conversion, combined with a novel DNA deaminase-helicase fusion that drives selective C-to-T conversion of natural cytosines in the original template strand. We demonstrate that Integrated-Seq is compatible with customizable enzymatic readouts to parse 5-methylcytosine and 5-hydroxymethylcytosine, and that preservation of the original four-base genetic code markedly improved enrichment, facilitating analysis of targeted genomic regions. Integrated-Seq thus provides a platform for simultaneous genetic and epigenetic analyses, paving the way for deep insights into fundamental biology and next-generation diagnostics.

  PublisherDOI

• Sparse-seq TAE Calculator

  Zenodo (CERN European Organization for Nuclear Research) · 2026-03-26

  otherOpen accessSenior author

  The TAE Calculator for Sparse-Seq, a method of quantifying cytosine modification using low-coverage sequencing https://zhou-lab.github.io/TAE_calculator/

  PublisherDOI

• Inducible, split base editors for in vivo cancer functional genomics

  Nature Biotechnology · 2026-04-15

  articleOpen access

  Cancer functional genomics using CRISPR base editors (BEs) holds great promise for molecular characterization and new target discovery. However, traditional BEs, using intact DNA deaminases as mutators, are often constrained by limited control and nonspecific toxicities. Here we developed a small-molecule-controllable system using split-engineered BEs (seBEs). By placing deaminase activity under small-molecule control, seBEs significantly reduced cellular toxicity and enabled robust and inducible in vivo functional genomics screens. High-density seBE genetic screens using ~11,000 single guide RNAs in vitro and ~3,700 single guide RNAs in vivo reveal known and previously unknown loss-of-function and dominant-negative mutations in cancer therapeutic targets. A deeper tiling seBE screen against Adar1, a key mediator in cancer immunotherapy, reveals critical residues within functional domains that show no phenotype in vitro but distinctively elicit non-cell-autonomous cancer dependencies in vivo. Overall, our seBE system offers a generalizable, controllable and highly efficient method to systematically identify key residues in cancer functional genomics.

  PublisherOA PDFDOI

• Abstract 1934 Genetic Code Expansion for Site-Specific Dual Encoding of Fluorophore-Quencher Pairs

  Journal of Biological Chemistry · 2025-05-01

  articleOpen access

  The ability to precisely modify proteins at multiple locations in their natural environment represents an unprecedented opportunity for answering biological questions at the molecular and cellular levels. In this work, we introduce a versatile dual incorporation approach, which involves the site-specific incorporation of two distinct noncanonical amino acids with bioorthogonal properties into proteins in vitro. This innovative strategy has demonstrated remarkable efficacy in incorporating acridone (Acd) and tetrazine (Tet) bearing amino acids at TAG and TAA codons in a single protein using mutually orthogonal tRNA/synthetase sets.

  PublisherOA PDFDOI

• Neomorphic leukemia-derived mutations in the TET2 enzyme induce genome instability via a substrate shift from 5-methylcytosine to thymine

  Proceedings of the National Academy of Sciences · 2025-01-28 · 7 citations

  articleOpen access

  Ten-eleven translocation (TET) enzymes oxidize 5-methylcytosine (mC) in DNA, contributing to the regulation of gene transcription. Diverse mutations of TET2 are frequently found in various blood cancers, yet the full scope of their functional consequences has been unexplored. Here, we report that a subset of TET2 mutations identified in leukemia patients alter the substrate specificity of TET2 from acting on mC to thymine. This neomorphic activity results from substitutions at key residues involved in the interactions with the mC base, including Asn1387 and His1904. Recombinant human TET2 proteins harboring the mutation of these residues can catalyze the oxidation of thymine to 5-hydroxymethyluracil (hmU) and 5-formyluracil (fU). Exogenous expression of the mutant TET2 Asn1387Thr (N1387T) in HEK293T cells leads to hmU accumulation, with levels further increased in cells lacking the glycosylase SMUG1. Endogenous knock-in of N1300T, the murine equivalent of N1387T, in mouse embryonic stem cells induces hmU production, causing DNA lesions and transcriptional activation of DNA damage response genes. N1300T cells accumulate more additional mutations with extended culture and exhibit heightened sensitivity to ATR inhibition compared to Tet2 knockout cells. Our study reveals that certain patient-derived TET2 mutations can acquire unexpected gain-of-function activities beyond impairing mC oxidation, offering a fresh perspective on the diverse molecular etiology of mutant TET2-related leukemogenesis.

  PublisherOA PDFDOI

• Renal Replacement Therapy

  2025-01-01

  book-chapterSenior author
  PublisherDOI

• Scalable screening of ternary-code DNA methylation dynamics associated with human traits

  Cell Genomics · 2025-07-03 · 7 citations

  articleOpen access

  Epigenome-wide association studies (EWASs) are transforming our understanding of the interplay between epigenetics and complex human traits. We introduce the methylation screening array (MSA) to enable scalable and quantitative screening of trait-associated DNA cytosine modifications in large human populations. The MSA integrates EWASs and cell-type-linked methylation signatures, covering diverse traits and diseases. Using the MSA to profile the ternary-code DNA methylations-dissecting 5-methylcytosine (5mC), 5-hydroxymethylcytosine (5hmC), and unmodified cytosine-revealed a previously unappreciated role of 5hmC in mediating human trait associations and epigenetic clocks. We demonstrated that 5hmCs complement 5mCs in defining epigenetic cell identities. In-depth analyses highlighted the cell-type context of EWAS and genome-wide association study (GWAS) hits. Targeting aging, we uncovered shared and tissue-specific 5hmC aging dynamics and tissue-specific rates of mitotic hyper- and hypomethylation. These findings chart a landscape of the complex interplay of the two forms of cytosine modifications in diverse human tissues and their roles in health and disease.

  PublisherDOI

• A ternary-code DNA methylome atlas of mouse tissues

  Genome biology · 2025-10-07 · 2 citations

  articleOpen access

  BACKGROUND: DNA cytosine modifications, including 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC), are key epigenetic regulators with distinct functions. Dissecting the ternary code (C, 5mC, 5hmC) across tissues and cell types remains a critical priority due to the limitations of traditional profiling methods based on bisulfite conversion. RESULTS: Here, we leverage the combined bisulfite and enzymatic (bACE) conversion with the Mouse Methylation BeadChip to generate 265 base-resolution ternary-code modification maps of 5mC and 5hmC across 29 mouse tissue types spanning 8-76 weeks of age and both sexes. Our atlas reveals a complex grammar of 5hmC distribution, jointly shaped by cell mitotic activity, chromatin states, and interplay with 5mC at the same and neighboring CpG sites. Of note, we demonstrate that 5hmC significantly complements 5mC-based biomarkers in delineating cell identity in both brain and non-brain tissues. Each modification state, including 5hmC alone, accurately discriminates tissue types, enabling high-precision machine learning classification of epigenetic identity. Furthermore, the ternary methylome variations extensively implicate gene transcriptional variation, with age-related changes correlated with gene expression in a tissue-dependent manner. CONCLUSIONS: Our work reveals how tissue, sex, and age jointly govern the dynamics of the two cytosine modifications, augments the scope of DNA modification biomarker discovery, and provides a reference atlas to explore epigenetic dynamics in development and disease.

  PublisherOA PDFDOI

• The bacterial stress response polymerase DinB tolerates sugar modifications and preferentially incorporates arabinosyl nucleotides

  RSC Chemical Biology · 2025-01-01

  articleOpen accessSenior author

  , translesion DNA synthesis can be fulfilled by Y-family DNA polymerases, including DNA polymerase IV (DinB). DinB features a more open active site and lacks proofreading ability, promoting error-prone replication. While DinB is known to tolerate damaged nucleobases like 8-oxo-guanine (8-oxoG), its ability to accommodate sugar-modified nucleotides has been underexplored, a question of importance given that such analogs are commonly used to inhibit viral and other polymerases. To explore DinB's selectivity, we screened a variety of sugar-modified noncanonical nucleotide triphosphates (nNTPs) and determined that DinB is intolerant of most 3'-modifications but can incorporate a subset of 2'-modifications. In particular, arabinosyl nucleotide triphosphates (araNTPs) showed efficient incorporation and limited extension. Furthermore, araNTPs can effectively compete with natural nucleotide triphosphates leading to stalled replication by DinB. We show that this tolerance extends to combined nucleobase and sugar modifications, with preferred misincorporation of 2'-fluoroarabinosyl-8-oxo-GTP opposite A more than C. Overall, our work highlights the potential for exploiting substrate promiscuity to target DinB and, thereby, slow bacterial adaptation to antibiotics.

  PublisherOA PDFDOI

Recent grants

• NIH Grant K08AI089242

  NIH · $660k · 2015

• The Molecular Basis for the Bacterial SOS Signal

  NIH · $1.2M · 2018–2023

• Ultra low-input epigenetic sequencing with combined enzymatic and long-read technologies

  NIH · $3.2M · 2019–2030

• APOBEC-Coupled Epigenetic Sequencing

  NIH · $443k · 2017–2019

• NIH Grant DP2GM105444

  NIH · $2.2M · 2017

Frequent coauthors

• Christopher T. Walsh

  Met Office

  27 shared

• Zachary M. Hostetler

  University of Pennsylvania

  20 shared

• E. James Petersson

  17 shared

• James T. Stivers

  Johns Hopkins University

  15 shared

• Jamie E. DeNizio

  University of Pennsylvania

  14 shared

• Mohamed A. Marahiel

  Philipps University of Marburg

  14 shared

• Robert F. Siliciano

  Howard Hughes Medical Institute

  14 shared

• Christian E. Loo

  University of Pennsylvania

  13 shared