Intro

Importance of protein interactions (PI)

  • understand disease and develop therapeutical targets

  • proteins translates the informations present in DNA into functions

  • PI implies molecular recognition

  • allows us to better understand pathologies and have a different approche than the traditional pathological pathway

  • more relevant in network medicine and § pharmacology

Study of interaction (only direct ones are wanted)

  • Build model of PI to make detection easier

  • Databases different dtb will results in different interactions depending on the technics used to identify interactions

    • IntAct

    • BioGRID

    • DIP

    • MINT

    • InnateDB

    • MatrixDB

    • IMex : mix of several dtb

    • String

  • It is better to try more than one dtb and CV the result

  • Biophysical analysis takes into consideration protein bindings so it has more chemico-physical components and it gives us the strength of the PI (equilibrium)

  • Structural Characterisation :

    • technics : X-Ray crystallography (pro: no size limit, con : crystallisation and its not exactly the same structure as real life as it requires stabilisation and purification), NMR (pro : possible in solution, con : size of the protein we can study depends on the magnet, it needs an additional step of structural modelling), Electron Microscopy (SAXS) (same resolution as X-Ray in some cases, pro : no crystallisation, con : cryogenic conditions and 2D is preferred)

    • Info gathered in various dtb including PDB, Dockground, Interactome3D, 3DComplex

Applications

  • Protein Design

  • Engineering protein - protein Interfaces : helps creating antibodies

  • Prediction and interpretation of pathological mutations

  • Drug Discovery

Computational methods

  • AlphaFold

    AlphaFold is a deep learning system developed by DeepMind that predicts protein folding structures. It uses AI to determine the 3D structure of a protein based on its amino acid sequence.

  • Needed to provide more structural (dynamic, conformational and interactions) models

  • Seq based technics and structural bioinformatics are used to modelize interaction

  • Interface prediction : ODA, NIP, OPRA, ET

  • Complex structure prediction (docking) :

    • Homology docking searches for differences in sequences Template Based

    • ab initio Docking (template free) :

      • Sequence based alignment : interactome3D

      • structure based alignment : used to identify templates, PRISM, Dockground, prePPI

    • AI docking