593. Optimizing interface evaluation in molecular domain
Invited abstract in session MB-56: CBBM: Drug Discovery, stream Computational Biology, Bioinformatics and Medicine.
Monday, 10:30-12:00Room: Liberty 1.11
Authors (first author is the speaker)
| 1. | Marta Szachniuk
|
| Institute of Computing Science, Poznan University of Technology | |
| 2. | Maciej Antczak
|
| Institute of Computing Science, Poznan University of Technology | |
| 3. | Olgierd Ludwiczak
|
| Poznan University of Technology |
Abstract
Operations research plays a critical role in optimizing computational methods for macromolecular structure analysis, particularly in predicting complex molecular structures and evaluating their accuracy. In the latter, traditional scoring often overlooks the quality of intermolecular interfaces, which are crucial for the accurate formation and function of molecular assemblies. In this presentation, we introduce Intermolecular Interaction Network Fidelity (I-INF), a normalized similarity measure quantifying intermolecular interactions in multichain complexes. Additionally, we implement the F1 metric to complement I-INF in the intuitive assessment of predicted 3D models, emphasizing interchain contacts. We validate I-INF and F1 on 72 RNA-protein decoys, demonstrating their effectiveness in ranking models and identifying high-quality predictions.
Keywords
- Computational Biology, Bioinformatics and Medicine
Status: accepted
Back to the list of papers