RUPEE: A fast and accurate purely geometric protein structure search

Abstract

Given the close relationship between protein structure and function, protein structure searches have long played an established role in bioinformatics. Despite their maturity, existing protein structure searches either use simplifying assumptions or compromise between fast response times and quality of results. These limitations can prevent the easy and efficient exploration of relationships between protein structures, which is the norm in other areas of inquiry. To address these limitations we have developed RUPEE, a fast and accurate purely geometric structure search combining techniques from information retrieval and big data with a novel approach to encoding sequences of torsion angles. Comparing our results to the output of mTM, SSM, and the CATHEDRAL structural scan, it is clear that RUPEE has set a new bar for purely geometric big data approaches to protein structure searches. RUPEE in top-aligned mode produces equal or better results than the best available protein structure searches, and RUPEE in fast mode demonstrates the fastest response times coupled with high quality results. The RUPEE protein structure search is available at https://ayoubresearch.com. Code and data are available at https://github.com/rayoub/rupee.

Computational structural genomics unravels common folds and predicted functions in the secretome of fungal phytopathogenMagnaporthe oryzae

“…We classified the 527 uncharacterized proteins by searching for similar structural folds and topologies in the SCOPe and CATH databases with RUPEE ( Fig. 1C; Table S3) (34)(35)(36). The classification identified enzymes of diverse functions.…”

“…The secondary structures and disordered residues were predicted with PSIPRED v4.0 and DISOPRED v3.16 (105,106). The predicted structures were compared with RUPEE against the SCOPe v2.07 and CATH v4.2.0 databases as well as the PDB chains (TOP_ALIGNED, FULL_LENGTH) (27,(34)(35)(36). Regardless of the confidence scores, a top hit was reported.…”

Protein structure search to support the development of protein structure prediction methods

“…Even when a large enough MSA is available, AlphaFold as well as traditional physics-based approaches to protein structure prediction, such as Rosetta, 6 have not reached the desired level of accuracy. 3 Previously, we introduced RUPEE, 7 a purely geometric protein structure search with no dependence on sequences or clustering.…”

“…Since the release of RUPEE, 7 we have observed that RUPEE has been used to upload protein structures that were the output of a protein structure prediction method in order to identify the most similar known structure to the predicted structures. For the most part, these uploaded protein structures have had low sequence and structure similarity to known structures in the PDB.…”

“…While the original search modes for RUPEE 7 have no dependence on sequences or clustering, they often lack sufficient sensitivity to find the most similar matches for a structure with low structure similarity to known structures. This lack of sensitivity for RUPEE fast and top-aligned search modes with respect to low similarity searches is due to the lower accuracy of their initial structure similarity estimates used to filter candidate matches.…”