Marc Pielies Avelli
PhD fellow
Research focus
Development and application of deep learning models in biomedical research. I’m mainly interested in using multimodal data integration methods and genomic language models to unveil the genomic mechanisms of disease.
Short profile
I studied theoretical physics at the University of Barcelona. During my bachelor’s I had the chance to take a couple of elective courses in medical physics and biophysics, which I really loved. I then enrolled in a masters programme in Biophysics and Computational Biology at Lund University, where I started to explore the use of AI on biomedical data. After my master’s I joined Simon’s lab as part of the last batch of the Copenhagen Bioscience PhD Programme (CBPP). I am now starting the last year of my PhD, which focuses on the development and application of deep learning approaches to tackle the variant-to-function (V2F) challenge.
Current projects:
- Modeling nascent transcription from chromatin landscape and structure
- On the use of VAEs for biomedical data integration.
- Unveiling genomic mechanisms of adipogenesis and insulin resistance with deep learning.
Future projects (open for collaborations):
- Building a V2F knowledge graph (GWAS, eQTLs, pQTLs, caQTLs, pathways).
Datasets (available):
- GTEx
- ENCODE genomic tracks
Toolkit:
Brief CV
- 9/2022–Present: PhD in Bioinformatics at the University of Copenhagen
- 9/2020–8/2022: MSc in Biophysics and Computational Biology at Lunds Universitet
- 9/2016–8/2020: BSc in Physics at the University of Barcelona
Other profiles
LinkedIn
Staff website at NNF CBMR
Keywords: Variant-to-function (V2F) studies, Variational Autoencoders (VAEs), DNA-based LLMs.