CENTRALE LYON - Post doctoral AI/ML for Clinical Proteomics (DIA Mass Spectrometry Data)

PROJECT OVERVIEW

We are looking for a highly motivated Postdoctoral Researcher to lead the development of innovative Artificial Intelligence (AI) and Machine Learning (ML) models for the analysis of Data-Independent Acquisition (DIA) proteomics data. This position is part of the ANR-funded

AIDIBOP project, which aims to design a novel, rapid diagnostic platform for Bloodstream Infections (BSI), capable of simultaneously identifying pathogens and determining their antibiotic resistance profiles directly from clinical samples.

You will play a key role within a multidisciplinary team (ISA Laboratory and Hospice Civil de Lyon), turning complex spectral data into meaningful clinical insights, to enhance patient outcomes and combat microbial resistance.

KEY RESEARCH AREAS AND RESPONSIBILITIES

You will be responsible for driving the core computational research of this project, which is structured around three key technical pillars:

1. Development of a State-of-the-Art Spectrum-Spectrum Matching (SSM) Model

• Develop SSM models for high-accuracy peptide identification from DIA data.

• Conduct a joint comparative study on feature selection and model architecture (e.g.,transformers) to maximize identification performance.

• Integrate spectral library generation tools into a robust, automated pipeline.

2. Protein Inference and Explainable AI (XAI) for Clinical Deployment

• Adapt existing protein inference algorithms to minimize false-negative rates, a critical metric for clinical diagnostics.

• Implement innovative XAI techniques (e.g., SHAP attention mechanisms) to provide peptide and protein-level explanations for model predictions, ensuring the pipeline is trustworthy and interpretable for healthcare professionals.

• Rigorously evaluate the pipeline using standard metrics (sensitivity, specificity, etc.).

3. Pioneering Direct Antibiotic Resistance Prediction from Spectral Data 1

• Lead the prospective task of developing a novel, sequencing-independent framework for direct antibiotic resistance prediction.

• Explore the transformation of MS/MS data into pseudo-images (MS1) or tensors (MS2) to leverage advanced Deep Learning architectures like CNNs.

• Compare the performance of this end-to-end approach against the traditional peptidebased pipeline, with a target of >80% sensitivity and specificity for Antimicrobial Susceptibility Testing (AST).

REQUIRED QUALIFICATIONS

• A Ph.D. in Computational Biology, Bioinformatics, Computer Science or Biostatistics.

• Expertise in AI/ML: Demonstrable, hands-on experience with machine learning and deep learning. Proficiency in Python and key libraries (e.g., PyTorch or TensorFlow, Scikitlearn, etc.) is mandatory.

• Research Mindset: A track record of innovation, problem-solving, and the ability to work independently on a complex, multi-faceted project.

• Software Skills: Experience with git version control and creating reproducible code.

• Excellent communication skills in English, with a proven ability to convey complex technical concepts to both specialist and non-specialist audiences.

ADDITIONAL SKILLS AND COMPETENCIES

• Experience with any of the following: Spectrum-Spectrum Matching models, transformer architectures, protein inference algorithms.

• Experience in handling and processing large-scale biological datasets.

• A background or strong interest in microbiology, infectious diseases, or microbial resistance.

WHY JOIN US

• A 2-year full-time postdoctoral position in a cutting-edge, clinically relevant project.

• The opportunity to publish high-impact papers and develop patentable methodologies.

• A monthly gross salary of 2,800 euros.

• A dynamic, collaborative, and supportive international research environment.

• Access to high-performance computing resources and high-quality, novel experimental datasets.

• An opportunity to see your contribution to the development of a life-saving diagnostic tool.

HOW TO APPLY

Please prepare a single PDF file containing:

1. A detailed Curriculum Vitae, including a list of publications.

2. A 1-page motivation letter explaining your specific interest in this project and how your background directly addresses the key responsibilities and qualifications listed above.

3. The names and contact information of at least two academic references