Dr Matteo Fumagalli

Profile

I graduated in biomedical engineering from Politecnico di Milano - Italy (PhD, 2011). I then moved to UC Berkeley for a postdoc position with Professor Rasmus Nielsen funded by EMBO. In 2014 I returned to Europe as a research fellow with Professor Francois Balloux at UCL funded by HFSP and jointly supervised by Professor Frances Brodsky with funding from NIH. I moved to Imperial College London in 2016 and then became Lecturer in Quantitative Evolution at Silwood Park in 2017. I joined Queen Mary University of London as Senior Lecturer in Genetics in 2021.

In my research, I use statistics and computer science to solve complex questions in human genetics and evolutionary biology. I am a strong advocate of open-source and open-access science. I often deliver outreach activities on human evolution at local communities and supervise secondary school pupils.

Undergraduate Teaching

Evolution (BIO113)

Essentials skills in biomedicine (BMD100)

Research projects (BIO600/BMD600)

Postgraduate Teaching

Statistics for Biologists (BIO724P)

Bioinformatics Software Development Group Project (BIO727P)

Research projects (BIO702P)

Research

Research Interests:

Has evolution shaped our predisposition to disease?

As humans encountered new environments, adaptation shaped the genetic similarities and differences among populations through the action of natural selection. In a changing environment, previously beneficial mutations may have a detrimental effect and contribute to disease susceptibility. We investigate how much natural selection has maintained disease-associated genetic variants in human populations, as a result of exposure of new pathogens, diet regimes, or extreme environments.

How do we extract reliable information from biological experiments?

Whilst new high-throughput sequencing technologies allow for rapid generation of genomic data, the information they provide is associated with uncertainty and errors. We develop statistical methods and implement software for the bioinformatic analysis of genomic sequencing data. We are particularly interested in estimating population genetic parameters from low-coverage sequencing data of non-model systems.

Is Artificial Intelligence able to solve the unsolvable in evolutionary studies?

One of the main scientific advancements in recent years is the improved predictive power of Artificial Intelligence. We pioneer the use of machine learning algorithms for the inference of past demographic history, admixture events, and genes under positive and balancing selection from genomic data. We established the EvoGenomics.AI for the dissemination of new findings, including a review in the application of Artificial Intelligence in evolutionary genomics.

Publications

Google Scholar publication list 

Supervision

Yuemin Li

Amelia Eneli

Collaborators

Michael DeGiorgio (Florida Atlantic University)

Sara Mathieson (Haverford College)

Austin Burt (Imperial College London)

Grants

NSFDEB-NERC: "Machine learning tools to discover balancing selection in genomes from spatial and temporal autocorrelations" 2023-2026

NERC Exploring the frontiers of environmental research: "Generative adversarial networks for demographic inferences of nonmodel species from genomic data" 2023

The Leverhulme Trust Research Grant: "A deep learning approach to quantify natural selection in Latin Americans" 2018-2022

Human Frontier Science Program postdoctoral fellowship 2014-2016

European Molecular Biology Organization postdoctoral fellowship 2012-2013