Ph.D. Candidate
Research Scientist @ Element AI
NSERC Alexander Graham Bell Scholar
Machine Learning and Computational Biology

Université Laval
Pavillon Adrien-Pouliot, Office 3908
1065, avenue de la Médecine
Québec (Québec), Canada G1V 0A6

Supervisor: François Laviolette
Co-supervisors: Mario Marchand , Jacques Corbeil
Research Groups: Groupe de Recherche en Apprentissage Automatique, Corbeil Research Group, Big Data Research Center


Main focus: Learning highly interpretable models for phenotype prediction in genomics
Machine learning algorithms can be used to correlate genomic variations (e.g., mutations, insertions, deletions) with biological states of interest (phenotypes). This setting is particularly challenging for learning algorithms, since the data is scarce and extremely high-dimensional. My research is oriented towards the development of algorithms that achieve strong regularization through sparsity and the use of prior knowledge, such as the evolutionary relationships that bind the learning examples. I am particularly interested in algorithms producing models that can be refined into knowledge that is valuable to domain experts.
Secondary topics:
  • Quality control of biological samples (e.g., blood donations) through machine learning and high-throughput mass-spectrometry
  • Algorithms for the alignment and correction of mass-spectrometry data
  • Group-based feature selection (e.g., biological pathways)

Peer-Reviewed Publications


Drouin, A., Giguère, S., Déraspe, M., Marchand, M., Tyers, M., Loo, V. G., Bourgault, A. M., Laviolette, F. & Corbeil, J. (2016). Predictive computational phenotyping and biomarker discovery using reference-free genome comparisons. BMC Genomics, 17(1), 754. [ pdf ]

Giguère, S., Drouin, A., Lacoste, A., Marchand, M., Corbeil, J., & Laviolette, F. (2013). MHC-NP: Predicting peptides naturally processed by the MHC. Journal of immunological methods, 400, 30-36. [ pdf ]

Giguère, S., Marchand, M., Laviolette, F., Drouin, A., & Corbeil, J. (2013). Learning a peptide-protein binding affinity predictor with kernel ridge regression. BMC bioinformatics, 14(1), 82. [pdf ]

Conferences and Workshops

Drouin, A., Hocking, T.D. & Laviolette, F. (2017). Maximum Margin Interval Trees. Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA. [ pdf ]

Drouin, A., Raymond, F., Letarte St-Pierre, G., Marchand, M., Corbeil, J. & Laviolette, F. (2016). Large scale modeling of antimicrobial resistance with interpretable classifiers. Machine Learning for Health Workshop, NIPS, Barcelona, Spain. [ pdf ]

Drouin, A., Giguère, S., Déraspe, M., Laviolette, F., Marchand, M. & Corbeil, J. (2015, July). Greedy Biomarker Discovery in the Genome with Applications to Antimicrobial Resistance. Greed is Great Workshop, ICML, Lille, France. [ pdf ]

Drouin, A., Giguère, S., Sagatovich, V., Déraspe, M., Laviolette, F., Marchand, M. & Corbeil, J. (2014, December). Learning interpretable models of phenotypes from whole genome sequences with the Set Covering Machine. Machine Learning in Computational Biology Workshop, NIPS, Montréal, Canada. [ pdf ]

Latulippe, M., Drouin, A., Giguère, P., & Laviolette, F. (2013, August). Accelerated robust point cloud registration in natural environments through positive and unlabeled learning. In Proceedings of the Twenty-Third international joint conference on Artificial Intelligence (pp. 2480-2487). AAAI Press. [ pdf ]


Invited Talks

Machine learning for antibiotic resistance: from rule-based models to deep architectures, Argonne National Laboratory, Argonne, Illinois (October 2017)

Interpretable Models of Antibiotic Resistance with the Set Covering Machine Algorithm, Google, Cambridge, Massachusetts (February 2017) [ pdf, video ]

Kover: un nouvel outil à base d’apprentissage automatique pour la découverte de biomarqueurs génomiques, Institut de Biologie Intégrative et des Systèmes, Québec, Canada (June 2016) [ pdf ]

Learning Extremely Sparse Classifiers from Whole Genomes Sequences with Set Covering Machines, Inria (SIERRA Team), Paris, France (February 2016) [ pdf ]

Genomic Biomarker Discovery with Set Covering Machines, Institut Curie, Paris, France (July 2015) [ pdf ]

Conferences and Workshops

Maximum Margin Interval Trees, Université Laval, Québec, Canada (May 2017) [ slides ]

Rule-based machine learning algorithms for antibiotic resistance prediction, GLBIO 2017, Chicago, Illinois (May 2017) [ slides ]

Large scale modeling of antimicrobial resistance with interpretable classifiers, Machine Learning for Health Workshop, NIPS, Barcelona, Spain. (December 2016) [ poster ]

Apprentissage de modèles à base de règles pour la prédiction de la résistance aux antibiotiques, Journées de la recherche en santé, Québec, Canada (May 2016) [ slides ]

Set Covering Machines and Reference-Free Genome Comparisons Uncover Predictive Biomarkers of Antibiotic Resistance, CCBC/GLBIO 2016, Toronto, Canada (May 2016) [ slides ]

Modelling Antibiotic Resistance with Sparse Machine Learning and Reference-Free Genome Comparisons, SMPGD 2016, Lille, France (February 2016) [ slides ]

Greedy Biomarker Discovery in the Genome with applications to Antibiotic Resistance, Greed is Great Workshop, ICML 2015, Lille, France (July 2015) [ slides ]

Apprentissage de modèles parcimonieux à partir de génomes complets avec application à la résistance aux antibiotiques, Université Laval, Québec, Canada (April 2015) [ slides ]


Introduction to machine learning, Microbiome Summer School 2017, Québec, Canada (June 2017) [ course material ]

Introduction to supercomputing with the Colosse supercomputer, Department of Computer Science and Software Engineering, Québec, Canada (May 2017) [ course material ]