Global informatics and physical property selection in protein sequences

Abstract

The degree of informatic independence between the physical properties of amino acids as encoded in actual protein sequences is calculated. It is shown that no physical property can be identified that carries significantly less information than others and that the information overlap between different properties and different length scales along the sequence is essentially zero. These observations suggest that bioinformatic models based on arbitrarily selected sets of physical properties are inherently deficient.protein bioinformatics | physical properties | information theory | Fourier analysis P rotein bioinformatics originated with the use of alignmentbased methods to compare protein sequences. In recent years, there has been a great increase in the use of "knowledgebased" methods, in which sequences are characterized by assigning to their component amino acids numerical values of physical properties that are believed to be important. These properties are usually selected to provide a quantitative basis for an intuitive picture of the physical chemistry of amino acids. Subsequent analysis is carried out using either a detailed description of the sequences of interest, which involves a consideration of local sequence characteristics, or a set of sequenceaveraged property values, which involves discussion of global sequence characteristics (1, 2).In previous work, we have considered various informatic aspects of the sequence-structure relationship in proteins. It was shown that there is structural uncertainty encoded in both local sequence information (3) and the global properties of sequences (4). These uncertainties constitute intrinsic limitations of knowledge-based protein bioinformatics. In this work, we examine investigator-imposed limitations resulting from the selection of "representative" sets of physical properties to characterize the amino acids. We ask whether an arbitrarily preselected set of amino acid physical properties can act as proxies for others that are not included. We also ask whether physical properties can be identified that are significantly less important than others and therefore, candidates for exclusion from knowledge-based models of the sequence-structure relationship. The following results are shown.i) The information-theoretic relationship between different global sequence physical properties can be computed. ii) Sequence-averaged and -specific variables can be treated in a unified manner. iii) When physical properties are expressed using an exhaustive and nonredundant representation, their longitudinal sequence distributions are found to be unrelated informatically. This observation implies that the use of a limited set of preselected variables is guaranteed to result in information loss. iv) The informatic contributions of all variables are found to be of the same order of magnitude. This fact implies that there are no variables that are safely negligible. ResultsOur approach is based on two tools. The first tool is the physical description of amino acids using the Property Fac...

Sequence‐specific dynamic information in proteins

“…The behavior summarized in Figure is qualitatively different from that observed when the 10 static property factors were compared in the same way . In that case, we found that all the significant sequence differences between architectural groups, at every level of the hierarchy, fall in the first bin (0 ≤ k ≤ 10).…”

“…If this is the case, we expect that groups of proteins that fold to different architectures will exhibit statistically significant differences in their dynamic propensities. We ask whether this is so, by addressing two specific questions: Are there indeed statistically significant differences in the characteristics of dynamic sequences between groups of proteins known to fold to different architectures? Is the behavior of those characteristics different from that we have observed for static properties? …”

“…We address the following goals: We propose an amino acid property that can act as an informatic basis for dynamic analysis. We investigate quantitatively the degree to which amino acid physical characteristics explain the values of this property. We use this property to develop a global dynamic representation of protein sequences, similar to that we developed in previous work for the static physical properties of protein sequences. This representation leads to a quantitative dynamic signature for the sequence. We determine the extent to which this representation encodes information previously unavailable. We use the new representation to investigate the dynamic differences between groups of proteins that fold to different architectures …”

Sequence-, structure-, and dynamics-based comparisons of structurally homologous CheY-like proteins

“…This representation carries 86% of the variance of the entire set of available physical properties (45,46). We have further shown (48) that the representation cannot be simplified, because any deviation from the full property factor representation results in a loss of physical information. No factor encodes information about any of the others, and all encode roughly equal amounts.…”

ProtDCal‐Suite: A web server for the numerical codification and functional analysis of proteins

“…In this context, ProtDCal is a software package that transforms protein sequences or 3D‐structures into general‐purpose numerical descriptors, accounting for both global and local information . Due to its complementary performance with respect to other well‐established tools in the field like PROFEAT and PseAcc (later extended to Pse‐in‐one), ProtDCal has been used in a number of studies . Notable among them are the modeling of posttranslational modifications, the prediction of protein enzymatic function, the prediction of antimicrobial activity in peptides, the determination of residues critical for protein function, and the prediction of stability changes upon mutations .…”