Construct a variable-length fragment library for de novo protein structure prediction

Abstract

Although remarkable achievements, such as AlphaFold2, have been made in end-to-end structure prediction, fragment libraries remain essential for de novo protein structure prediction, which can help explore and understand the protein-folding mechanism. In this work, we developed a variable-length fragment library (VFlib). In VFlib, a master structure database was first constructed from the Protein Data Bank through sequence clustering. The hidden Markov model (HMM) profile of each protein in the master structure database was generated by HHsuite, and the secondary structure of each protein was calculated by DSSP. For the query sequence, the HMM-profile was first constructed. Then, variable-length fragments were retrieved from the master structure database through dynamically variable-length profile–profile comparison. A complete method for chopping the query HMM-profile during this process was proposed to obtain fragments with increased diversity. Finally, secondary structure information was used to further screen the retrieved fragments to generate the final fragment library of specific query sequence. The experimental results obtained with a set of 120 nonredundant proteins show that the global precision and coverage of the fragment library generated by VFlib were 55.04% and 94.95% at the RMSD cutoff of 1.5 Å, respectively. Compared with the benchmark method of NNMake, the global precision of our fragment library had increased by 62.89% with equivalent coverage. Furthermore, the fragments generated by VFlib and NNMake were used to predict structure models through fragment assembly. Controlled experimental results demonstrate that the average TM-score of VFlib was 16.00% higher than that of NNMake.

Protein Multiple Conformation Prediction Using Multi-Objective Evolution Algorithm

“…MultiSFold is a multi-objective population optimization method based on multiple distance information, focusing not only on the accuracy of the final structure but also on the generation of multiple models with diversity that may reflect conformational transitions. In addition to the query sequence, the variable-length fragment library generated by our previously proposed method VFlib (Feng et al, 2022) and several inter-residue distance maps are required as input. The initial population is generated by randomly fragment assembly.…”

“…In this work, the input variable length fragment library will be first generated based on the query sequence and the corresponding MSAs through our previously developed VFlib (Feng et al, 2022). Then, several input distance maps generated by different predictors and a consensus distance map based on them will be used to guide the conformation sampling.…”

Adaptive Sampling Methods for Molecular Dynamics in the Era of Machine Learning

“…This resembles early approaches to structure prediction, such as Rosetta. 74 A more recent exploration of this approach was presented in MultiSFold, 69 a method that combines the variable-length fragment library (VFLib) 75 technique with multiple structure prediction models to diversify the predicted conformational ensemble. This method is particularly useful when different prediction models (e.g., AlphaFold and RosettaFold) predict different structures for the same protein.…”