Embedded Files
Nonribosomal Peptide Synthetases
Nonribosomal Peptide Synthetases
US 9,487,763 B2 · Granted November 8, 2016
Application: US 14/713,662
Inventors: David H. Sherman, Michael M. Kaufman-Schofield, Sunit Jain, and Gregory J. Dick
The patent covers the biosynthetic pathway and polynucleotide sequences involved in producing ET-743 (trabectedin), a natural-product cancer drug. I developed the computational approach used to identify the responsible gene cluster from environmental metagenomic data. The work also resulted in a co-first-author publication in Environmental Microbiology.
View on Google Patents
Strain Abundance Estimation Technology
Strain Abundance Estimation Technology
US Patent Application 17/999,516
Technology for strain-level abundance quantification in complex microbial communities. The associated computational method, NinjaMap, supported the design and evaluation of a defined synthetic human microbiome and underwent FDA IND review as part of that program.
Related publication: Cheng et al., “Design, construction, and in vivo augmentation of a complex gut microbiome,” Cell (2022).
Microbiome-Based Immuno-Oncology
Microbiome-Based Immuno-Oncology
US Patent Application 17/420,813
Research arising from computational and experimental work on microbiome-derived immuno-oncology candidates at Second Genome. My contributions included multi-modal meta-omics analysis and computational screens used to evaluate and prioritize therapeutic hypotheses.
Novel Endonuclease Protein Design
Novel Endonuclease Protein Design
Patent application filed by Acrobat Genomics
AI-assisted design of novel endonuclease variants. I developed an ESM3-based strategy combining deep-mutational-scan data, literature-derived constraints, conservation scores, ESMFold structure prediction, and RMSD-based filtering. Of 32 candidates selected from approximately 36,000 computational designs and tested experimentally, four outperformed wild type in the functional assay.
A public patent number is not yet listed.
These inventions reflect a consistent research approach: use computation to search biological space at a scale that would otherwise be inaccessible, then reduce that space to hypotheses and designs that can be tested experimentally.
Report abuse