Institut de Mathématiques de Marseille, UMR 7373


Accueil > Séminaires > Séminaires et Groupes de travail hebdomadaires > Mathématiques et Algorithmique pour la Biologie des Systèmes (MABioS)

Séminaire Mathématiques et Algorithmique pour la Biologie des Systèmes (MABioS)

par Baudot Anaïs, Lozingot Eric, Remy Elisabeth, Tichit Laurent - publié le , mis à jour le



  • Lundi 29 avril 11:00-12:00 - Francisco Pinto - Univ. Lisboa

    Predicting protein annotations using the expected network interaction between related diseases or cellular processes

    Résumé : Genes or proteins involved in the same cellular process or disease tend to cluster together within biological networks. This fact is used by numerous algorithms to predict new protein functions or disease associations.
    We hypothesize that knowing that two (or more) cellular processes or diseases are functionally related may improve our ability to predict new proteins associated with both processes or diseases. I will present two methods that are being developed in my research group based in the previous hypothesis.
    The first is a specific version of betweenness centrality, called S2B, that predicts the simultaneous association of a protein with two related diseases. For each candidate protein, this method measures the frequency with which that protein is part of shortest paths linking proteins associated with one disease to proteins associated with the other disease.
    The second method aims to expand the list of proteins associated with a seed cellular process. It first identifies other cellular processes that specifically interact with seed proteins of the queried cellular process. Seed direct neighbors are then compared with the seeds by their ability to bridge the query process with the specific neighbor processes. This comparison allows the classification of seed direct neighbors as being involved (or not) in the seed process.

    Exporter cet événement

  • Lundi 13 mai 11:00-12:00 - Quentin Ferré - TAGC & LIS

    Correlating TRBS ChIP-seq peaks from multiple datasets in ReMap using deep convolutional autoencoders

    Résumé : Cis-regulatory elements (CREs) are genomic regions regulating gene expression by binding proteins called Transcriptional Regulators (TRs). TR binding is mostly studied experimentally, via ChIP-Seq, but these experiments have false positives, and there is no method to discern them. However, TRs are known to be co-occurent, and many replica datasets exist. As such, we use common TR and/or datasets combinations to identify “atypical” peaks. We use the ReMap database to learn such correlations.
    CREs are represented as 3D tensors of peak presence (namely ‘position’, ‘TR’, and ‘dataset’). We use an autoencoder to perform a lossy compression of each, to keep common patterns and discard rare elements (atypical peaks). The regions are viewed by the model through convolutional filters to focus on the correlations. Each peak gets an anomaly score corresponding to the autoencoder reconstruction error.
    We use artificial data to confirm the model’s ability to discover correlation groups of TR/datasets and label lonely/anomalous peaks. Application to ReMap is in progress, currently on a curated subset of data. To our knowledge, our research shows the first use of a large-scale meta-analysis to corroborate different ChIP-Seq datasets, using deep learning to integrate them in complex combinations and eliminate atypical peaks.

    Notes de dernières minutes :

    Exporter cet événement

  • Lundi 17 juin 15:00-16:00 - Maxime Lucas - IBDM, Marseille

    Séminaire Mathématiques et Algorithmique pour la Biologie des Systèmes (MABioS)

groupe de travail

Manifestation scientifique

Nature Séminaire et Groupe de Travail
Intitulé Mathématiques et Algorithmique pour la Biologie des Systèmes (MABioS)
Responsables Elisabeth Remy & Laurent Tichit
Équipe de rattachement BioMath Alea (BMA) du Groupe ALEA
Fréquence 1 séance chaque semaine
Jour-Horaire Lundi, de 14h à 16h
Lieu Group Meetings : salle de réunion 3ème étage
Séminaires : amphithéâtre Herbrand
I2M, Luminy (accès)

Contact :