DescFold

DescFold is a software tool for fold recognition that uses machine learning-based methods. It combines four descriptors using Support Vector Machines (SVMs). The descriptors include a profile-sequence-alignment-based descriptor using Psi-blast e-values and bit scores, a sequence-profile-alignment-based descriptor using Rps-blast e-values and bit scores, a descriptor based on secondary structure element alignment (SSEA), a descriptor based on the occurrence of PROSITE functional motifs.

This work focuses on the improvement of DescFold by incorporating more powerful descriptors and setting up a user-friendly web server. In seeking more powerful descriptors, the profile-profile alignment score generated from the COMPASS algorithm was first considered as a new descriptor (i.e., PPA). When considering a profile-profile alignment between two proteins in the context of fold recognition, one protein is regarded as a template (i.e., its 3D structure is known). Instead of a sequence profile derived from a Psi-blast search, a structure-seeded profile for the template protein was generated by searching its structural neighbors with the assistance of the TM-align structural alignment algorithm.

Moreover, the COMPASS algorithm was used again to derive a profile-structural-profile-alignment-based descriptor (i.e., PSPA). The new DescFold was trained and tested on a total of 1,835 highly diverse proteins extracted from the SCOP 1.73 version. When the PPA and PSPA descriptors were introduced, the new DescFold boosted the performance of fold recognition substantially.

The DescFold method was benchmarked against several well-established fold recognition algorithms through the LiveBench targets and Lindahl dataset and demonstrated very competitive performance. It can serve as a useful tool to assist in template-based protein structure prediction. To provide a large-scale test for the new DescFold, a stringent test set of 1,866 proteins was selected from the SCOP 1.75 version. At a less than 5% false positive rate control, the new DescFold correctly recognized structural homologs at the fold level for nearly 46% of test proteins.