PIGSPro: prediction of immunoGlobulin structures v2

Area of Science:

• Computational biology
• Structural bioinformatics
• Immunology

Background:

• Accurate prediction of immunoglobulin structure is crucial for understanding immune responses and developing therapeutics.
• Existing methods like the PIGS server have limitations in prediction accuracy and user interface.

Purpose of the Study:

• To introduce PIGSpro, a significantly upgraded server for predicting immunoglobulin structure.
• To enhance the accuracy and usability of immunoglobulin structure prediction through algorithmic and interface improvements.

Main Methods:

• Rewritten pipeline in Python incorporating updated homology modeling, template alignment, and canonical structure models for hypervariable loops.
• Advanced prediction of the heavy chain H3 loop and improved packing of antibody chains.
• Redesigned user interface and automated monthly database updates.

Main Results:

• Demonstrated improvements in prediction accuracy across various steps of the immunoglobulin structure prediction pipeline.
• Successful integration of updated computational methods developed over the years.
• Enhanced user experience through a completely redesigned interface.

Conclusions:

• PIGSpro represents a substantial advancement in computational tools for immunoglobulin structure prediction.
• The server offers improved accuracy and accessibility for researchers in immunology and drug discovery.
• Continuous updates ensure the PIGSpro database remains current with the latest advancements.