Sequence and intramolecular distance scoring analyses of microbial rhodopsins

Abstract

Recent accumulation of sequence and structural data, in conjunction with systematical classification into a set of families, has significantly advanced our understanding of diverse and specific protein functions. Analysis and interpretation of protein family data requires comprehensive sequence and structural alignments. Here, we present a simple scheme for analyzing a set of experimental structures of a given protein or family of proteins, using microbial rhodopsins as an example. For a data set comprised of around a dozen highly similar structures to each other (overall pairwise root-mean-squared deviation < 2.3 Å), intramolecular distance scoring analysis yielded valuable information with respect to structural properties, such as differences in the relative variability of transmembrane helices. Furthermore, a comparison with recent results for G protein-coupled receptors demonstrates how the results of the present analysis can be interpreted and effectively utilized for structural characterization of diverse protein families in general.

Evaluation of variability in high-resolution protein structures by global distance scoring

“…Each of the panels is a “main plot” for a protein and the pattern represents the current status of a protein’s 3D structural variation depicted in a 2D figure; this reflects the overall fold and its variability/invariability shown in the lower/upper part of the plot, respectively. The initial purpose of this plot was to delineate the common features in rhodopsin-like GPCRs [7] ; then, the usefulness of it was further evaluated with another family of 7TM helical proteins, microbial rhodopsins [9] . Based on these studies, we are confident that the plot can be used for any proteins to concisely grasp crystallographically observed structural variations.…”

“…A modified version of a Python script (score-analyzer [9] , sa_v16) performs the following functions: reading a PDB entry, extracting/displaying of the C α coordinates (choosing one from the alternative conformations, if present), calculating and storing distance data per chain for a selected sequence range, calculating average scores for making a “progress plot”, extracting and assessing the high-scoring populations in a defined set of distance ranges, and plotting and saving graphs and tables. Additional functions for GPCRs are implemented to associate the general amino acid position numbers (BW numbers) [24] with each C α and to classify the C α –C α pairs (e.g., intrahelical or interhelical, Fig.…”

“…For example, intramolecular distances between all C α pairs in a polypeptide of 100 amino acids generate 100 × 99/2 = 4950 data whereas the number of RMSDs at C α s is equal to only 100 data. Our previous studies on GPCRs and other 7TM proteins [9] showed that an inverse of “coefficient of variation [standard deviation (stdev) divided by average]”, assigned to the distance of each C α pair, appeared as a comprehensive numeral and reasonably reflected the observed variability/invariability. Thus, we term it a “score” and the method as distance scoring analysis (DSA).…”

Sequence and intramolecular distance scoring analyses of microbial rhodopsins

“…F1000Research : Dataset 1. Raw data for DSA ( Figure 2 – Figure 6 , Figure S3 – FigureS6 ), 10.5256/f1000research.7920.d113285 22 …”

“…F1000Research : Dataset 2. Python script for making a score vs distance plot, 10.5256/f1000research.7920.d113889 23 …”