Mots-Clés

Bioinformatics Deep learning protein design host-pathogen interaction

Description

Context

Antimicrobial resistance (AMR) is one of the top ten global public health threats facing humanity and is predicted to cause 10 million deaths yearly by 2050. A high rate of resistance against commonly used antibiotics have been increasingly observed world-wide, showing an essential need for designing novel therapeutics to increase the effectiveness of fighting pathogens. Accordingly, the detailed understanding of the molecular interactions between pathogens and their hosts is crucial.

The main goal of this PhD project is to study the surface interactome of an important human-specific pathogen in interaction with human/mouse plasma to elucidate the protein-protein interactions that help the bacteria to evade the immune responses. Such information will be used then to design protein binders to inhibit such interactions. To achieve this goal, the candidate will study the surface virulence factors to find best targets, and design specific binders in silico for selected targets which can be further validated in vitro.

The candidate will be hosted in the CAPSID team, LORIA, Inria, Nancy Grand Est. The candidate will be supervised by Hamed Khakzad (Inria Junior Professor) with expertise in integrative structural biology, host-pathogen interactions, protein design, and deep learning [1,2,3,4]. The team consists of several permanent researchers with expertise in macromolecular interactions and docking, structural biology, and deep learning, together with several PhD and master students.

 

Assignment

Background: Streptococcus Pyogenes (Group A Streptococcus; GAS) is an important human-specific antibiotic-resistant pathogen causing both mucosal and systemic infections. It produces several secreted and surface-attached virulence factors to target host proteins, localizing and initiating infections. GAS has evolved multiple immune camouflage strategies including scavenging host proteins by its virulence factors to build a dense layer of protein-protein interactions (PPI) on its surface [1]. The surface interactome of GAS is largely unknown reflected by the large number of uncharacterized proteins in its proteome. This lack of knowledge is one of the major limitations in therapeutic/vaccine design strategies against GAS.

Main task: This PhD position aims to provide a comprehensive picture of the surface interactome of GAS including the interactome of its major virulence factors by using integrative structural biology approaches. Such interactome has been previously addressed for the M1 protein, the most important GAS antigen through several integrative approaches combining cross-linking mass spectrometry, cryo-EM, and electron microscopy [1, 4, 5, 6]. Here, the goal is to provide computational predictions for a large set of interactions and incorporate the obtained knowledge to design specific inhibitory proteins toward those targets.

References

1. S. Hauri and H. Khakzad et al., "Rapid determination of quaternary protein structures in complex biological samples," Nature Communications, vol. 10, no. 192, 2019.
2. JK. Leman et. al., “Macromolecular modeling and design in Rosetta: recent methods and frameworks,” Nature Methods, vol. 17, no. 7, 2020
3. C. Goverde et. al., “De novo protein design by inversion of the AlphaFold structure prediction network,” biorxiv, 2023.
4. H. Khakzad et al., "Structural determination of Streptococcus M1 protein interaction with human IgGs using targeted cross-linking mass spectrometry," PLoS Computational Biology, vol. 17, no. 1, p. e1008169, 2021.
5. L. Happonen et al., "A quantitative Streptococcus pyogenes–human protein–protein interaction map reveals localization of opsonizing antibodies," Nature Communications, vol. 10, no. 2727, 2019.
6. S. Chowdhury et al., "Streptococcus pyogenes Forms Serotype- and Local Environment- Dependent Interspecies Protein Complexes," mSystems, vol. 6, no. 5, 2021.

Main activities

1. Implementing large-scale methods and preparing a software using Python
2. Validating the method and analysing the results
3. Computational design de novo proteins
4. Collaborating with microbiology and protein design teams.
5. Writing scientific articles and presenting the work in international conferences

Skills

• Master degree in Computer Science, Bioinformatics, Chemoinformatics or a related program
• Proficiency in programming languages (Python) and good coding practices is mandatory
• Skills in protein design is a plus
• Experience in machine learning and/or deep learning (scikit, PyTorch)
• Ability to work independently and also to work in a team
• Excellent oral and written English skills

Benefits package

• Fully funded position
• Subsidized meals
• Partial reimbursement of public transport costs
• Professional equipment available
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage
• More information about Inria: https://www.inria.fr/en/inria-centre-universite-lorraine

Duration

Duration: 36 months

Application

Send a CV and a motivation letter to Hamed Khakzad: hamed.khakzad@inria.fr