A new definition and properties of the similarity value between two protein structures

Abstract

Knowledge regarding the 3D structure of a protein provides useful information about the protein's functional properties. Particularly, structural similarity between proteins can be used as a good predictor of functional similarity. One method that uses the 3D geometrical structure of proteins in order to compare them is the similarity value (SV). In this paper, we introduce a new definition of the SV measure for comparing two proteins. To this end, we consider the mass of the protein's atoms and concentrate on the number of protein's atoms to be compared. This defines a new measure, called the weighted similarity value (WSV), adding physical properties to geometrical properties. We also show that our results are in good agreement with the results obtained by TM-SCORE and DALILITE. WSV can be of use in protein classification and in drug discovery.

Persistent homology reveals strong phylogenetic signal in 3D protein structures

“…This issue requires further investigation. To limit the biases, it has been proposed to remove the amino acids corresponding to indels from structures before calculating distances ( 17 ) or to normalize distances ( 82 , 83 ). A key question is whether to include indels or leave them out, as most studies based on sequence comparison do.…”