Here’s a slightly more in-depth look at some of my technical projects. Or, you can visit my Google Scholar page.
- CovCysPredictor Targeted covalent inhibition is a powerful therapeutic modality in the drug discoverer’s toolbox. However, identifying cysteines for targeted covalent inhibition is a difficult task, as experimental and in silico tools have shown limited accuracy. We release CovCysPredictor, a simple, interpretable machine learning model for predicting ligandable cysteines. For more information, see the full paper recently published at JCIM. (January 2025)
- Combinatorial CRISPR Screen The interaction between genetics and the environment (a so-called GxE interaction) is well-appreciated, but sometimes the body encodes ‘redundant’ proteins that perform the same function, and these are very difficult to assess the functionality of using traditional GxE approaches. In this study, we looked at GxGxE interactions to elucidate the mechanisms of different mitochondrial solute carriers using a combinatorial CRISPR screen. You can read more in our Nature Communications paper (May 2022)
- Biomarkers of Mitochondrial Disease Severity To identify robust circulating markers of m.3243A>G disease, we first performed discovery proteomics, targeted metabolomics, and untargeted metabolomics on plasma from a deeply phenotyped cohort (102 patients, 32 controls). In a validation phase, we measured concentrations of prioritized metabolites in an independent cohort using distinct methods. We validated 20 analytes (1 protein, 19 metabolites) that distinguish patients with MELAS from controls. Read more in our full paper published in JCI (January 2021)
- STRATA: Adversarial Attacks for Models of Code STRATA is a technique for generating effective, low-cost (gradient-free) adversarial examples for models of code (AdvML KDD ‘21). (Summer 2020)
- Grammatical Evolution This project implements an extensible grammatical evolution framework that allows for various types of grammatical evolution contests, including Prisoner’s Dilemma and the Hawk-Dove game, with spatial variants included. (Winter 2020)
-
Natural Isotope Correction The natural abundance of
heavyelements can sometimes skew experiments based on labelled isotopes. This package corrects for the natural abundance of heavy elements to allow for good inference in certain types of chemistry experiments. Related to the work for this paper (Winter 2018)