Soutenance d’Habilitation à Diriger des Recherches, Bio-informatique, 28/04/2016

J’ai soutenu mon Habilitation à Diriger des Recherches (HDR), intitulée “Contributions à l’élaboration de connaissances qualitatives en bio-informatique“, le jeudi 28 avril 2016 à 10h30 dans l’amphithéâtre S de l’Institut de Recherche en Communications et Cybernétique de Nantes (IRCCyN) devant le jury composé de :

Rapporteurs :
Gilles BERNOT, PU, Université de Nice Sophia Antipolis (I3S)
François FAGES, DR Inria, Inria Paris-Saclay
Cédric LHOUSSAINE, PU, Université de Lille 1 (LIFL)

Examinateurs :
Jérémie BOURDON, PU, Université de Nantes (LINA)
Hidde DE JONG, DR Inria, Inria Grenoble-Rhône Alpes
Denis THIEFFRY, PU, École normale supérieure (Institut de biologie)

Présidente :
Mireille RÉGNIER, DR Inria, Inria Paris-Saclay & École Polytechnique (LIX)

Garant HDR :
Olivier F. ROUX, PU, École centrale de Nantes (IRCCyN)

Manuscrit

Lien vers le manuscrit “Contributions à l’élaboration de connaissances qualitatives en bio-informatique“.

Vidéo

Retransmission de la Soutenance d’Habilitation à Diriger des Recherches, Bio-informatique, 28/04/2016, Dr Morgan Magnin

Slides

Lien vers les slides projetées lors de la soutenance.

Résumé

Cette synthèse pour l’HDR est un recueil de plusieurs travaux dans le domaine des méthodes formelles pour la biologie. Devant l’enjeu que représente la modélisation et l’analyse de la dynamique de réseaux de régulation biologiques à grande échelle, nous avons identifié une classe pertinente de modèles formels sur laquelle il est possible de mener des analyses efficaces de la dynamique. Après avoir discuté les différents critères de modélisation à prendre en compte en biologie, nous introduisons ainsi le formalisme des Frappes de Processus. Nous présentons ensuite les méthodes d’analyse conçues pour ce paradigme, leurs mérites et leurs limites. Enfin, nous revenons sur des résultats plus récents, consécutifs à l’établissement de liens fructueux entre l’apprentissage automatique, la programmation logique, le model-checking et la bio-informatique, ce qui nous permet de faire émerger un ensemble de nouvelles questions scientifiques.

Mots clés
réseaux de régulation biologique, modélisation, frappes de processus, analyse de la dynamique, apprentissage automatique, vérification formelles, inférence de paramètres

Summary

I defended my habilitation thesis, entitled “Contributions to the elaboration of qualitative knowledge in bioinformatics“, on Thu. 2016/04/28. This synthesis gathered several works in the field of formal methods for biology. In our works, we tackled the challenge of modeling and analyzing the dynamics of large-scale biological regulatory networks. To address this issue, we identified some relevant class of formal models on which it is possible to perform effective analysis of its dynamics. After discussing the different modeling criteria to be taken into account in biology, we introduce the Process Hitting framework. We then present the methods that we designed to analyze such models, and their respective merits and limitats. Finally, we give an overview of recent research aiming to build a fruitful link between machine learning, logic programming, model-checking and bioinformatics. This allows us to bring out a new set of scientific questions.

Keywords
biological regulatory networks, modeling, process hitting, analysis of the dynamics, machine learning, model-checking, parameters inference

Propositions de sujets de Master 2 en informatique pour l’année 2015-2016

À l’occasion de l’année universitaire 2015-2016, je propose trois différents sujets de Master, chacun étant en co-encadrement avec l’un de mes collègues. Ces travaux se tiendront à l’IRCCyN, à Nantes. Pour postuler sur l’un de ces sujets, merci de contacter les personnes mentionnées dans le descriptif détaillé des sujets, que vous pourrez retrouver ci-dessous :

Day 2 & 3 at ILP’2015 (25th Conference on Inductive Logic Programming)

IMG_0497

Here is a follow-up of my previous summary about the 25th Conference on Inductive Logic Programming that I have just attended in Kyoto. The current post is not self-consistent, in the sense that it should be read alongside with my summary of day 1 of ILP’2015 (to illustrate it, the numbering of this post is a continuation of the former one).

I will focus here on day 2 and day 3, to conclude with general considerations about my feelings about this community and the organization of the conference. As a reminder for those willing to know more, organizers uploaded all papers on ILP website.

2015/08/img_1415.jpg

6/ Morning of day 2 focused on probabilistic approaches in learning. Taisuke Sato, who gave a talk about the distribution semantics he proposed 20 years ago, insisted on his conviction that probabilistic logic learning and statistical relational learning will play a major role for machine learning techniques to be able to tackle huge and noisy data, and he showed how his distribution semantics is a way to address such issues.

2015/08/img_1672.jpg

7/ The third – and last – invited talk, in day 3, from Luc de Raedt, also discussed probabilistic approach, more specifically probabilistic logic programming. From ProbLog (which is a probabilistic Prolog), various applications were given, from the identification of the magnetic (or not) nature of objects based on observation to biological analysis of the connection between genes in a graph.

8/ A significant number of talks emphasized the need to benefit from modern computational structure (cloud computing, clusters, …) to make the ILP approaches be able to address large-scale data. As a consequence, it is necessary to develop specific algorithms that are able to take profit from the new available infrastructures. Different methods are proposed, like the use of a MapReduce approach, specific contributions to use GPU kernels or alternative parallel-processing systems.

9/ In some applications (e.g., Luc de Raedt presented a nice and challenging case study for a robot to recognize graspable points), it is crucial to process relational kernels on graphs. Approaches around kernels arouse much efforts from the community.

10/ I have been struck by the wide range of applications of ILP approaches. While ILP, and more generally logics, is still considered as a difficult topic that many bachelor or master students fear, it definitely has many practical, useful and elegant applications. Obviously systems biology, in which I have a special interest, but also robotics (e.g., rescue robots), medicine (to analyze the correlations between medical records and illnesses), object recognization, … All these topics are really exciting, as they question our capacity to abstract the world in terms of facts, rules and more generally logics.

2015/08/img_1485.jpg

11/ This session of ILP ended on a panel celebrating the 25th anniversary of the conference. The chairs of the five last editions shared their views about the current trends and next outcomes for ILP. Many topics and keywords about it were discussed. Among them are the emergence of new application fields (Stephen Muggleton for example quoted synthetic ecology), the huge challenge of deep learning, or the improvement of parallel-techniques to guarantee ILP techniques can process real-life applications on a daily basis.

12/ As a first-time attendent to ILP, I was deeply interested in the quality of the talks. The mix between conceptual researches and practical illustration of the merits of such works makes the conference to be accessible.

13/ As a final word, the conference’s lunches and social dinner has contributed to make it memorable. Tasty bento, and a dinner that allowed everyone to experience being served by maikos and listen to classical shamisen dances and songs. This really was what the Japanese call omotenashi (おもてなし), i.e., the sense of hospitality.

Day 1 at ILP’2015 (25th Conference on Inductive Logic Programming)

IMG_1354

First day of ILP’2015 in Kyōto is over. This is the first time I attend the conference. It is a nice dive in this community!

The (scientific) facts that I will keep in mind from this first day:
1/ Meta-interpretive learning is trendy by its ability to tackle predicate invention (3 talks about it). It has promising applications for robot strategies, proof strategies and so on.

2/ Behind these “meta” interests, there is a question of designing systems that automatically find, produce and reason over consistent axiomatic systems.

3/ Biology provides challenging scientific questions for inductive logic programming, in terms of analysis of pathways, regulations or structure of components. Among relevant questions are the inference/completion of models through logical approaches, with the need to address noisy data.

4/ I liked the prospective talk about “robot engineer”, consisting in the use of ILP techniques to produce a system that, given an action to be performed, automatically builds the robot that will satisfy/perform such goal.

5/ Finally, on a pure organizational aspect, I enjoyed the fact that all papers are available online, being able to have an overview to all of them before and during the talks.

I hope tomorrow will as interesting as today!

Research activity in July 2014

Gorgeous view on the Japanese Imperial Palace and Garden, from one of the meeting rooms at the National Institute of Informatics, Tôkyô

July has been my second month as JSPS fellow in National Institute of Informatics, Tôkyô. I have continued to work on different collaborations here with Japanese researchers, both on bioinformatics (especially in the learning of gene regulatory networks) and systems resilience.

International (peer-reviewed) conferences

I got two new accepted papers:

  • T. Ribeiro, M. Magnin, and K. Inoue. Learning delayed influence of dynamical systems from interpretation transition. To appear in the 24th International Conference on Inductive Logic Programming. 6 pages (short paper). Nancy, France, September 2014.In a nutshell: This paper addresses the learning of (synchronous) logical programs with delayed influences from state transition diagrams. Delayed influences are captured with an inductive logic programming methodology.
    Note: This paper results from my collaboration with Inoue Lab. at NII.
  • S. Carolan, M. Magnin and A-L. Kabalu. Sparking a Digital. Revolution: Digital Educational Tools in Fragile and Emerging Learning Contexts. In the 1st International Conference dedicated to Digital Society and Cultures (DI’2014). Nantes, France, September 2014.In a nutshell: This paper discusses the crucial issue of teaching and learning for people in hard environments (e.g. warzones, countries with poor or broken Internet connections, etc.) and gives insights about the added-value of recent e-Learning methods.
    Note: This paper results from joint work with Simon Carolan (2nd year PhD student who I co-supervise) and Anne-Laure Kabalu (Master student who Simon and I have co-supervised this Spring).

Students’ talks

Meanwhile, the PhD and Master students I co-supervise have presented some recent works at some international conferences and research schools: