PROFcon: novel prediction of long-range contacts

Area of Science:

• Computational biology
• Structural bioinformatics
• Protein structure prediction

Background:

• The increasing disparity between known protein sequences and experimentally determined structures necessitates advanced prediction methods.
• Current prediction tools partially address the sequence-structure gap, highlighting the need for improved accuracy in predicting residue contacts.
• Accurate prediction of non-local residue contacts is crucial for enhancing comparative modeling, fold recognition, and experimental structure determination.

Purpose of the Study:

• To introduce PROFcon, a novel computational method for predicting contacts between amino acid residues in proteins.
• To evaluate the performance of PROFcon using various data sources and compare it with existing state-of-the-art methods.

Main Methods:

• PROFcon integrates information from multiple sources: evolutionary sequence alignments, predicted secondary structure and solvent accessibility, inter-residue sequence separation, and overall protein properties.
• The method was tested on diverse protein families, including those with sparse evolutionary profiles and different SCOP classifications.

Main Results:

• PROFcon demonstrated comparable accuracy for both short and long proteins.
• Prediction accuracy was highest for proteins in the SCOP alpha/beta class and was reduced for protein families with few homologs (sparse evolutionary profiles).
• PROFcon outperformed other state-of-the-art methods in contact prediction accuracy at the CASP6 meeting, advancing the field's capabilities.

Conclusions:

• PROFcon represents a significant advancement in protein contact prediction, pushing the boundaries of accuracy.
• The method's predictions are sufficiently accurate to guide the prediction of global protein structural features.
• PROFcon's improved contact prediction capabilities can aid in understanding protein structure and function.