CDD de 12 mois de postdoc/engineer : “Méthode de prédiction de résistance aux antibiotiques”, Centre de Bioinformatique de Bordeaux (33)

Proposition et implémentation d’un approche Intelligence Artificielle pour la prédiction de résistance aux antibiotiques à partir de données génomiques et métabolomiques

Offre publiée sur le site Indeed.com le 05 mars 2021

This offer is in the context of an international projects centred on prediction of AntiMicrobial Resistance (AMR). We aim to develop machine learning methods that can orient the selection of treatments by assessing the level of resistance, provide rational for the generation of novel antibiotics, and assist in the surveillance of human and livestock AMR around the globe. To achieve this, we have assembled a transnational team (Canada, China, Finland, France) with complementary skills with demonstrated expertise in machine learning applied to both genomics and metabolomics and AMR domain experts.

We will focus initially on two important AMR pathogens, Pseudomonas aeruginosa (Pa) and Streptococcus pneumoniae (Spn). Indeed, Pa is a Gram negative, Spn a Gram positive; they have different genome sizes (Pa, 6Mb, Spn, 2Mb); different GC contents (Pa, GC rich, Spn AT rich); and Pa has more secondary metabolites than Spn. While both species have large pan-genomes, Pa has large extrachromosomal elements while in Spn, AMR is often linked to the formation of mosaic genes. Indeed, the mode of AMR transmission differs with Pa relying mostly on conjugation and Spn on transformation, although both acquire mutations to achieve AMR. Thus, ML strategies and computer derived decision trees developed and validated for those two distinct pathogens should be applicable to a broad range of additional AMR pathogens.

We posit that, in the early implementation phase of this change in the paradigm for AMR detection, validation of the ML algorithms will require experimental molecular reconstruction of resistance pathways. Indeed, biomarkers of AMR (being DNA sequences or metabolites) can be associated with AMR but not necessarily directly involved and this must be studied experimentally. Importantly, members of our team can expertly manipulate the Pa and Spn genomes for testing the role of SNVs, CNV, genes, and metabolites in AMR. Moreover, our goal is not only to determine whether a bacterium is sensitive or not to an antibiotic but also provide actionable information concerning the minimal inhibitory concentration (MIC) determination. The MIC value provides information on the level of resistance and orients treatment.

We already have generated genomic data for hundreds of Pa and Spn strains together with detailed resistance (MIC) information and matabolomic data is on the way.

The project of the post-doc / engineer focus on innovation in modelling and predicting resistance to improve diagnostics and its implementation.

Durée du contrat : 12 mois

Type d’emploi : Temps plein, CDD

Mesures COVID-19:
Port du masque, télétravail

Langue:

• Anglais (Obligatoire)

Télétravail:

• Temporairement en raison du COVID-19

Précautions contre le COVID-19:

• Processus de recrutement à distance
• Consignes de distanciation sociale

Pour postuler : candidater sur le site Indeed.com.