Machine Learning for Single-Cell Biology
Swiss computational biologist with expertise in machine learning, genomics, single-cell transcriptomics, large-scale bioinformatics pipelines, and AI-driven biomedical discovery. Over 4 years of professional experience as a Senior Data Scientist and Bioinformatician.
10+
Years Research
6
Computational Methods
20+
Scientific Publications
AI
Drug Discovery & Omics
Research Focus
Machine Learning Applied to Single-Cell Data
Development of advanced deep learning models for cell-type identification, transcriptomics integration, and biological state prediction.
Large-Scale Omics Analysis
Expertise across scRNA-seq, bulk RNA-seq, CLIP-seq, ChIP-seq, ATAC-seq, SNV calling, and epigenetics datasets.
Scalable Infrastructure
Extensive experience with Slurm, SCITAS, TACC, cloud infrastructure, Nextflow workflows, and distributed computing.
Professional Experience & Education
Senior Data Scientist | Alithea Genomics, Switzerland
Leading the Pipeline Development Team, AI-guided toxicity and mechanism-of-action prediction research, project management, Scrum organization, and R&D data analysis.
Ph.D. in Computational Biology | EPFL
Research focused on machine learning for transcriptomics, spatial biology, and deconvolution of drug screening data.
Biostatistician & Bioinformatician | EPFL
Developed algorithms for RNA-binding site detection and enrichment analysis of non-coding genomic features.
Teaching Assistant | EPFL
Leader of the teaching assistant team for the Object Oriented Programming C++ course.
M.S. / B.S. Life Sciences & Engineering | EPFL + UT Austin
Developed deep learning approaches for spatial transcriptomics and integration of single-cell RNA-seq with ex-vivo drug sensitivity data.
Developed Computational Methods
CLIMB
Bulk deconvolution method for estimation of cell-type fractions and expression in bulk transcriptomic samples.
CLIFF
Deconvolution framework for ex-vivo drug sensitivity datasets into cell-type specific drug responses.
TLCpeaks
Predictive peak-calling algorithm for RNA-binding proteins and CLIP-seq experiments.
pyTEnrich
Statistical enrichment framework for transposable element analysis in ChIP-seq data.
MatISSe
Autoencoder-like deep learning model for cell-type identity prediction in spatial transcriptomics.
alTErego
Inference of transposable element-mediated regulons and regulatory programs.
Selected Publications
Deconvolution of ex-vivo drug screening data and bulk tissue expression predicts the abundance and viability of cancer cell subpopulations
BioRxiv • 2023Detection and benchmarking of somatic mutations in cancer genomes using RNAseq data
PeerJ • 2018DPPA2 and DPPA4 are necessary to establish a 2C-like state in mouse embryonic stem cells
EMBO Reports • 2019Additional Co-Authorships
Publications in Science Advances, Cell Stem Cell, Mobile DNA, Genome Research, Nature Communications, Genome Biology, and BioRxiv.
Technical Skills & Activities
Programming Languages
Machine Learning
Expertise ranging from classical linear models to advanced deep learning architectures using PyTorch and scikit-learn.
Teaching & Conferences
Teacher of Single-Cell Data Analysis at BC2 Conference 2023, co-founder of the EDCB Symposium at EPFL, and participant in AMLD, SingleCellGenomics, and Cell Symposia.