Human Peptidyl-prolyl cis-trans isomerase NIMA-interacting 1 (hPin1) AlphaFold predicted structures

Description

This dataset is a collection of AlphaFold predicted structures of putative transient intermediates during the translation of hPin1. The structures predicted span residues 1-39, focusing on the folding of the tryptophan-tryptophan (WW) domain 5-39. The sequences used in all the structure prediction include the first 4 N-terminal residues ("MADE") and their nomenclature works as follows. Each model is named "made_", which indicates the leading residues not part of the WW domain, followed by "ww" and the progressive number of residues part of the WW domain. For example, the sequence "MADE", not including any of the WW domain residues, is "made_ww00", whereas the model corresponding to the longest sequence (amino acids 1-39), whcih includes all 35 residues part of the WW domain, is "made_ww35". Folder and file names are followed by "s" and a number referring to predictions obtained using the specific seeds 1, 10 and 100; for example "made_ww00s1" was obtained using the sequence "MADE" and a seed of 1.

Steps to reproduce

The amino acid sequence of hPin1 was obtained from UniProt (Q13526). A series of 36 intermediates, from fragment 1–4 up to 1–39, was generated by elongating the sequence one residue at a time, and each sequence was submitted to the AlphaFold server to predict the corresponding structure. For every sequence, predictions were run three times with fixed random seeds (1, 10, and 100) using the default PDB template set (cut off date 30/09/2021). For extra calculations beyond the raw output from AlphaFold server and reported in dataset_summary.xlsx, such as the secondary structure allocation via DSSP and the calculation of the RMSD from a reference structure, details are given in the associated publication.

Institutions

• University of Lincoln

  England, Lincoln

Categories

Licence