Modelling signaling on the single-cell level • Understanding Signal Flow • Fast & memory efficient software
──────────────
I am building fast, memory-efficient software to study signaling on the single-cell level. Leveraging novel single-cell datasets like sequencing based single-cell protein measurement technologies I am working on scalable tools to decode cellular signaling networks. My work spans the full analytical pipeline, from raw data processing and normalization to downstream network analysis of single-cell datasets (with focus on phospho-protein measurements). I develop tools from raw data processing to tools that study signaling on the single-cell level. E.g., I developed ESGI — Efficient Splitting of Generic Indices — a generic tool to processes barcoded single-cell data fast and memory efficient to go from raw FASTQ-files to your single-cell feature matrices and runs easily on standard laptops at home. Other tools include LoCo - Local Correlations - and R-package to detect local correlation patterns in single-cell data that are cell-state specific. LoCo is based on a high-performance C++ framework to calcualtes in seconds correlations in thousands of single-cell neighbourhoods across hundrets of features.
I develop fast, memory-efficient, and scalable software solutions for single-cell data analysis, with a strong focus on extracting meaningful biological insights from complex datasets. My current research centers on (phospho)protein data, where I study correlation patterns at the single-cell level to infer signaling networks. My work spans the full analytical pipeline, from raw data processing and normalization to downstream network analysis of single-cell protein datasets. I develop tools that process data in ways that enables truthful recovery of co-variation pattern between proteins. Subsequently I study how these features co-vary and this information (encoded in the mutual variation of proteins) flows through signaling networks and thereby defines how cells reponde to drug-treatment, external factors, etc. In the future I plan to combine these mechanistic pathway modelling techniques (as interpretable network backbones) with statistical machine learning methods to generalize singnaling on the single-cell level with biologically grounded interpretations. Doing so I hope to develop toosl that help engineer cell-fate behavior, study how to push cells into vulnerable states, predict their response to treatment or enhance productivity like biomass/ protein or aromatic compound production.
I am proficient in C++, R, and Python, and particularly interested in performance-oriented computing and scalable algorithms design for large and complex biological datasets.
I am based in Leiden, just finished writing my PhD thesis (at the NKI and VU Amsterdam) and am looking now for new opportunities. I previously studied Biotechnology at TU Berlin and Bioinformatics at the University of Hamburg.
────────────────────────────────────────────
Click on below projects for information how to de-barcode your single-cell data or unravel local correlation patterns────────────────────────────────────────────
ESGI - Efficient Splitting of Generic Indices:
LoCo - Local Correlation Analysis:
