PACT - Prediction of amyloid cross-interaction by threading.

Area of Science:

• Biochemistry and Molecular Biology
• Computational Biology and Bioinformatics
• Structural Biology

Background:

• Amyloid proteins are implicated in various diseases, including Alzheimer's and Parkinson's.
• Functional amyloids play crucial roles in physiological processes like biofilm formation.
• Cross-interactions between amyloids from different species can influence aggregation pathways and disease progression.

Purpose of the Study:

• To develop a computational method for predicting amyloid cross-interactions.
• To investigate the potential influence of microbial amyloids on human protein aggregation.
• To provide a tool for high-throughput studies of amyloid interactions.

Main Methods:

• Development of PACT (Prediction of Amyloid Cross-interaction by Threading), a computational method.
• PACT models heterogeneous fibrils formed by two amyloidogenic peptides.
• Structural statistical potential is used to assess the plausibility and stability of modeled fibrils.

Main Results:

• PACT achieved high prediction accuracy (AUC=0.88, F1=0.82) on the AmyloGraph database.
• PACT predicted cross-interactions between bacterial CsgA and human alpha-synuclein (Parkinson's disease) and hIAPP (type 2 diabetes).
• Experimental validation confirmed CsgA fragment interactions with hIAPP.

Conclusions:

• PACT is an effective computational tool for predicting amyloid cross-interactions.
• The method can identify specific protein regions involved in these interactions.
• PACT facilitates research into amyloid diseases and the role of microbial amyloids.