Laurent Jacob

Presentation

I am a currently a Genentech postdoctoral innovation fellow in the department of statistics of UC Berkeley.

I recently finished my Ph.D. at the École des Mines de Paris, in the Center for Computational Biology lead by Jean-Philippe Vert. The final manuscript is available here.

My research deals with statistical learning, and its application to biological and therapeutic problems, in particular cancer outcome prediction, molecular data analysis and drug discovery.

Education

• 2009 - Ph.D. at the École des Mines de Paris.
• 2006 - Master MVA (Mathematics, Vision, Learning) at the École Normale Supérieure de Cachan.
• Fall 2004 - One semester at Universidad Politécnica de Valencia.
• 2000-2005 - Université de Technologie de Compiègne ("Grande École"). Major: Computer Sciences, Minor: Philosophy, Technology and Cognition.
• 2000 - Lycée Hélène Boucher (Paris), Scientific Baccalauréat with honors.

A more complete resume can be downloaded here.

Publications

• Short research summary
• BibTeX

2010

• Increasing stability and interpretability of gene expression signatures. A.-C. Haury, L. Jacob, J.-P. Vert. Techreport, 2010.
  [arXiv]

2009

• Structured priors for supervised learning in computational biology. L. Jacob. Ph.D. dissertation, 2009.
  [Manuscript][Defense slides]
• Group Lasso with Overlap and Graph Lasso. L. Jacob, G. Obozinski, J.-P. Vert. International Conference on Machine Learning (ICML 26), 2009.
  [Link][Slides][Video][Code and data]
• Clustered Multi-Task Learning: A Convex Formulation. L. Jacob, F. Bach, J.-P. Vert. Advances in Neural Information Processing Systems (NIPS 21), 2009.
  [Link][Old techreport (arXiv)]

2008

• Virtual screening of GPCRs: an in silico chemogenomics approach. L. Jacob, B. Hoffmann, V. Stoven, J.-P. Vert. BMC Bioinformatics, 9:363, 2008.
  [Link] [Old techreport (arXiv)]
• Protein-ligand interaction prediction: an improved chemogenomics approach, L. Jacob and J.-P. Vert. Bioinformatics, 24(19):2149-2156, 2008.
  [Link] [Old techreport (arXiv)] [NIPS 2007 MLCB workshop slides] [CAp 2008 conference slides]
• Machine learning for in silico virtual screening and chemical genomics: new strategies, J.-P. Vert and L. Jacob. Combinatorial Chemistry & High Throughput Screening, 11(8):677-685, 2008.
  [Link] [Old techreport]
• Efficient peptide-MHC-I binding prediction for alleles with few known binders, L. Jacob and J.-P. Vert. Bioinformatics, 24(3):358-366, 2008.
  [Link] [Application (KISS)] [Old techreport (arXiv)]

Before 2008

• Master report, Multitask learning for epitope prediction.
  [Report] [Defense slides]
• Final project for Université de Technologie de Compiègne, Développement d'outils de clustering pour l'analyse de données en génomique fonctionnelle (in french).
  [Report] [Poster]

Teaching

• Spring 2008, "Apprentissage en bioinformatique et noyaux pour structures" for the machine learning module of École des Mines de Paris (in french).
  Lecture slides, material and questions for the practical session.
• Spring 2007, "Apprentissage en bioinformatique et noyaux pour structures" for the machine learning module of École des Mines de Paris (in french).
  Lecture slides (inspired by Jean-Philippe Vert's slides and Pierre Mahé's PhD defense slides), material for the practical session.