Eshita Mutt

Improved Detection of Remote Homologues Using Cascade PSI-BLAST: Influence of Neighbouring Protein Families on Sequence Coverage

Background Development of sensitive sequence search procedures for the detection of distant relationships between proteins at superfamily/fold level is still a big challenge. The intermediate sequence search approach is the most frequently employed manner of identifying remote homologues effectively. In this study, examination of serine proteases of prolyl oligopeptidase, rhomboid and subtilisin protein families were carried out using plant serine proteases as queries from two genomes including A. thaliana and O. sativa and 13 other families of unrelated folds to identify the distant homologues which could not be obtained using PSI-BLAST. Methodology/Principal Findings We have proposed to start with multiple queries of classical serine protease members to identify remote homologues in families, using a rigorous approach like Cascade PSI-BLAST. We found that classical sequence based approaches, like PSI-BLAST, showed very low sequence coverage in identifying plant serine proteases. The algorithm was applied on enriched sequence database of homologous domains and we obtained overall average coverage of 88% at family, 77% at superfamily or fold level along with specificity of ∼100% and Mathew’s correlation coefficient of 0.91. Similar approach was also implemented on 13 other protein families representing every structural class in SCOP database. Further investigation with statistical tests, like jackknifing, helped us to better understand the influence of neighbouring protein families. Conclusions/Significance Our study suggests that employment of multiple queries of a family for the Cascade PSI-BLAST searches is useful for predicting distant relationships effectively even at superfamily level. We have proposed a generalized strategy to cover all the distant members of a particular family using multiple query sequences. Our findings reveal that prior selection of sequences as query and the presence of neighbouring families can be important for covering the search space effectively in minimal computational time. This study also provides an understanding of the ‘bridging’ role of related families.

Proteome profile of a pandemic Vibrio parahaemolyticus SC192 strain in the planktonic and biofilm condition

Vibrio parahaemolyticus is one of the leading causative agents of foodborne diseases in humans. In this study, the proteome profiles of the pandemic strain V. parahaemolyticus SC192 belonging to the O3:K6 serovar during the planktonic and biofilm stages were analyzed by two-dimensional liquid chromatography coupled to tandem mass spectrometry. This non-gel-based multidimensional protein identification technology approach identified 45.5% of the proteome in the reference genome V. parahaemolyticus RIMD 2210633. This is the largest proteome coverage obtained so far in V. parahaemolyticus and provides evidence for expression of 27% of the hypothetical proteins. Comparison of the planktonic and biofilm proteomes based on their cluster of orthologous groups, gene ontologies and KEGG pathways provides basic information on biofilm specific functions and pathways. To the authors’ knowledge, this is the first study to generate a global proteome profile of the pandemic strain of V. parahaemolyticus and the method reported here could be used to rapidly obtain a snapshot of the proteome of any microorganism at a given condition.

LenVarDB: database of length-variant protein domains

Protein domains are functionally and structurally independent modules, which add to the functional variety of proteins. This array of functional diversity has been enabled by evolutionary changes, such as amino acid substitutions or insertions or deletions, occurring in these protein domains. Length variations (indels) can introduce changes at structural, functional and interaction levels. LenVarDB (freely available at http://caps.ncbs.res.in/lenvardb/) traces these length variations, starting from structure-based sequence alignments in our Protein Alignments organized as Structural Superfamilies (PASS2) database, across 731 structural classification of proteins (SCOP)-based protein domain superfamilies connected to 2 730 625 sequence homologues. Alignment of sequence homologues corresponding to a structural domain is available, starting from a structure-based sequence alignment of the superfamily. Orientation of the length-variant (indel) regions in protein domains can be visualized by mapping them on the structure and on the alignment. Knowledge about location of length variations within protein domains and their visual representation will be useful in predicting changes within structurally or functionally relevant sites, which may ultimately regulate protein function. Non-technical summary: Evolutionary changes bring about natural changes to proteins that may be found in many organisms. Such changes could be reflected as amino acid substitutions or insertions–deletions (indels) in protein sequences. LenVarDB is a database that provides an early overview of observed length variations that were set among 731 protein families and after examining >2 million sequences. Indels are followed up to observe if they are close to the active site such that they can affect the activity of proteins. Inclusion of such information can aid the design of bioengineering experiments.

Rapid and enhanced remote homology detection by cascading hidden Markov model searches in sequence space

MOTIVATION In the post-genomic era, automatic annotation of protein sequences using computational homology-based methods is highly desirable. However, often protein sequences diverge to an extent where detection of homology and automatic annotation transfer is not straightforward. Sophisticated approaches to detect such distant relationships are needed. We propose a new approach to identify deep evolutionary relationships of proteins to overcome shortcomings of the available methods. RESULTS We have developed a method to identify remote homologues more effectively from any protein sequence database by using several cascading events with Hidden Markov Models (C-HMM). We have implemented clustering of hits and profile generation of hit clusters to effectively reduce the computational timings of the cascaded sequence searches. Our C-HMM approach could cover 94, 83 and 40% coverage at family, superfamily and fold levels, respectively, when applied on diverse protein folds. We have compared C-HMM with various remote homology detection methods and discuss the trade-offs between coverage and false positives. AVAILABILITY AND IMPLEMENTATION A standalone package implemented in Java along with a detailed documentation can be downloaded from https://github.com/RSLabNCBS/C-HMM SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online. CONTACT mini@ncbs.res.in.

Search for Protein Sequence Homologues that Display Considerable Domain Length Variations

Independent folding units which have the capability of carrying out biological functions have been classified as “protein domains†. These minimal structural units lead not only to considerable sequence changes of protein domains of similar folds and functions, but also gives rise to remarkable length variations under evolutionary pressure. Rapid and heuristic sequence search algorithms are generally sensitive and effective in recognizing protein domains that are distantly related within large sequence databases, but are not well-suited to identify remote homologues of varying lengths. An even more challenging aspect is introduced to distinguish reliable hits from a vast number of putative false positives that could have suboptimal sequence similarities. Here, the authors present a data-mining approach that provides stage-specific filters in sequence searches to reliably accumulate remote homologues, which encourages sampling of length variations albeit with a low false positive rate. Realization of such remote homologues with vivid length variations could contribute to better understanding of functional variety within protein domain superfamilies.

Graph theoretic approach for studying correlated motions in biomolecules

The paper describes the application of graph theoretic concepts to the dynamic cross-correlation data obtained from MD simulations of adenine riboswitch, in the absence and presence of adenine. This novel approach combines both community detection algorithms that support edge weights, and cliques. The effect of variations in the values of nearest neighbors (NN) and correlation coefficient threshold (T) in the community detection algorithm have been applied to identify and filter out coincidental correlations between rogue nodes. The results generated for add Adenine riboswitch based on this hybrid approach, successfully identified the correlations within the structural regions of the molecule, providing strong clues regarding the functionality and stability of the RNA molecule in the absence and presence of adenine. Our results also suggested that a prior application of the proposed algorithm (in an automated fashion) to the simulation data of RNA biomolecules, can provide strong leads for hypothesis formulation and subsequent hypothesis-driven manual investigation.

Molecular Dynamics Simulations and Structural Analysis to Decipher Functional Impact of a Twenty Residue Insert in the Ternary Complex of Mus musculus TdT Isoform.

Insertions/deletions are common evolutionary tools employed to alter the structural and functional repertoire of protein domains. An insert situated proximal to the active site or ligand binding site frequently impacts protein function; however, the effect of distal indels on protein activity and/or stability are often not studied. In this paper, we have investigated a distal insert, which influences the function and stability of a unique DNA polymerase, called terminal deoxynucleotidyl transferase (TdT). TdT (EC:2.7.7.31) is a monomeric 58 kDa protein belonging to family X of eukaryotic DNA polymerases and known for its role in V(D)J recombination as well as in non-homologous end-joining (NHEJ) pathways. Two murine isoforms of TdT, with a length difference of twenty residues and having different biochemical properties, have been studied. All-atom molecular dynamics simulations at different temperatures and interaction network analyses were performed on the short and long-length isoforms. We observed conformational changes in the regions distal to the insert position (thumb subdomain) in the longer isoform, which indirectly affects the activity and stability of the enzyme through a mediating loop (Loop1). A structural rationale could be provided to explain the reduced polymerization rate as well as increased thermosensitivity of the longer isoform caused by peripherally located length variations within a DNA polymerase. These observations increase our understanding of the roles of length variants in introducing functional diversity in protein families in general.

Structural updates of alignment of protein domains and consequences on evolutionary models of domain superfamilies

BackgroundInflux of newly determined crystal structures into primary structural databases is increasing at a rapid pace. This leads to updation of primary and their dependent secondary databases which makes large scale analysis of structures even more challenging. Hence, it becomes essential to compare and appreciate replacement of data and inclusion of new data that is critical between two updates. PASS2 is a database that retains structure-based sequence alignments of protein domain superfamilies and relies on SCOP database for its hierarchy and definition of superfamily members. Since, accurate alignments of distantly related proteins are useful evolutionary models for depicting variations within protein superfamilies, this study aims to trace the changes in data in between PASS2 updates.ResultsIn this study, differences in superfamily compositions, family constituents and length variations between different versions of PASS2 have been tracked. Studying length variations in protein domains, which have been introduced by indels (insertions/deletions), are important because theses indels act as evolutionary signatures in introducing variations in substrate specificity, domain interactions and sometimes even regulating protein stability. With this objective of classifying the nature and source of variations in the superfamilies during transitions (between the different versions of PASS2), increasing length-rigidity of the superfamilies in the recent version is observed. In order to study such length-variant superfamilies in detail, an improved classification approach is also presented, which divides the superfamilies into distinct groups based on their extent of length variation.ConclusionsAn objective study in terms of transition between the database updates, detailed investigation of the new/old members and examination of their structural alignments is non-trivial and will help researchers in designing experiments on specific superfamilies, in various modelling studies, in linking representative superfamily members to rapidly expanding sequence space and in evaluating the effects of length variations of new members in drug target proteins. The improved objective classification scheme developed here would be useful in future for automatic analysis of length variation in cases of updates of databases or even within different secondary databases.

Mining Protein Sequence Databases for Remote Homologues That Can Display Considerable Domain Length Variations

Protein domains are minimal structural units that can independently fold and carry out discrete biological functions. Evolutionary divergence amongst proteins not only cause considerable sequence changes of protein domains of similar folds and functions, but can also give rise to remarkable length variations. Rapid and heuristic sequence search algorithms are generally sensitive and effective in recognizing protein domains that are distantly related within large sequence databases, but are not well-suited to identify remote homologues of varying lengths. It is also challenging to distinguish reliable hits from a vast number of putative false positives that could have sub optimal sequence similarities. Here, we present a data-mining approach that provides stage-specific filters in sequence searches to reliably accumulate remote homologues which encourages sampling of length variations albeit no compensation on the validity of hitherto identified distant relationships. Realization of remote homologues with vivid length variations could contribute to better understanding of functional variety within protein domain super families.