Career Profile

I work on developing machine learning methods for safety quantification of CRISPR/Cas9 gene editing in the group of Dr. Minary. We combine ab initio molecular simulation with data science and artificial intelligence algorithms to elucidate dangerous off-target effects in RNA-mediated gene editing. My research is embedded into an interdisciplinary training cohort (DTP), involving constant collaboration and communication with peers across scientific disciplines.

I have a strong background in the physical sciences, having worked in the field of ultracold atomic gases at the Universities of Cambridge, Durham and Heidelberg before.

Please do not hesitate to contact me using the email address in the sidebar.


piCRISPR: Physically Informed Features Improve Deep Learning Models for CRISPR/Cas9 Off-Target Cleavage Prediction

Recent years have seen continuous development of off-target prediction algorithms trained on in vitro cleavage assay data gained from immortalised cell lines. Here, we implement novel deep learning algorithms and feature encodings for off-target prediction and systematically sample the resulting model space in order to find optimal models and inform future modelling efforts. We find that our best-performing model highlights the importance of sequence context and chromatin accessibility for cleavage prediction and outperforms state-of-the-art prediction algorithms in terms of area under precision-recall curve.

crisprSQL: a novel database platform for CRISPR/Cas off-target cleavage assays

Even though a multitude of off-target cleavage datasets have been published, a comprehensive, transparent overview tool for CRISPR off-target data has not yet been established. Here, we present crisprSQL (, an interactive and bioinformatically enhanced collection of CRISPR/Cas9 off-target cleavage studies aimed at enriching the fields of cleavage profiling, gene editing safety analysis and transcriptomics.


These are a few of the projects which I have completed to date, both academic and extracurricular.

Protein Network Analysis - An investigation into the topological dimension of protein-protein interaction networks (PPINs).
Hybrid Trap Loading - A simulation package for atom transport in a combined magnetic and optical dipole trap.
Counterdiabatic Driving Simulations - A comprehensive simulation of atom transport using a single-beam optical dipole trap (optical tweezer), based on the semiclassical truncated Wigner approximation.
Abiforen - An online platform for the graduating years of high schools to assist with their data collection for year books etc.

Skills & Proficiency

(pandas, scipy, pytorch)





Javascript & jQuery