Structure comparison software tools | Protein data analysis
Quantitative comparison of two structures of the same biological macromolecule or complex is a very common but by no means trivial task. One such example is the comparison of a model with the experimentally determined (reference) structure.
Allows users to study the proteomic scale inference of enzyme function. EFICAz can identify functionally discriminating residue (FDR) as residues that discriminate the members of a homo-functional family from a hetero-functional family. It combines the prediction from four independent methods, namely: (1) CHIEFc family-based (FDR) identification, (2) multiple PFAM-based FDR recognition, (3) CHIEFc SIT evaluation and (4) high-specificity multiple PROSITE patterns.
Predicts functional annotations. AnnoLite annotates the structures processed by DBAli. The web application searches the DBAli database for structurally similar proteins and collects their known annotations. It annotates the input protein structure with CATH and SCOP fold assignments, EC numbers, InterPro entries, PFAM families and Gene Ontology (GO) and codes. The software can annotate a structure for all of the functional properties, with the exception of the GO cellular component term.
Predicts ligand-binding sites as well as protein interaction patches on the surface of the query structure. AnnoLyze is a web application that annotates the structures processed by DBAli using the DBAli, LigBase, PiBase and ModBase databases. The software annotates a target structure based on its stored relationships to other structures.
Serves to resolve problem of distant-homology proteins in X-ray crystallography. I-TASSER-MR produces full-length models working with I-TASSER by iterative structural fragment reassembly. Then, it uses a progressive sequence truncation procedure for editing the models based on local variations of the structural assembly simulations. This tool permits users to decide between different methods for setting B-factors and the number of models used for phasing.
Aims to build an accurate alignment for the entire length of a target sequence from a set of alternative inputs. MMM brings a method that assists users to identify low-sequence regions of “twilight” and “midnight zones". It detects the alternatively aligned regions from a set of input alignments, and the best scoring regions from a set of alternative segments are combined with the core part of the alignments to produce the final MMM alignment.
Allows global comparison and research for small dissimilar fragments. MCQ4Structures applies mean of circular quantities (MCQ) for the purpose of structure similarity identification. It is based on trigonometric representation of the molecule 3D structure and the inter-molecular distance measured as MCQ. This tool allows to aggregate over any number of differences computed for pairs of corresponding angles. The software also implements the LCS-TA algorithm, that finds fragments displaying high similarity in torsion angle space.
A server and tools designed for largescale substructure comparison. The LabelHash Server predicts the function of unknown proteins. The algorithm consists of two stages: (i) a preprocessing stage (the algorithm builds up lookup tables for n-tuples of residues that occur in a set of target structures) and (ii) a matching stage (the algorithm first looks for partial matches of size n and incrementally expands these partial matches to complete matches in a depth-first fashion).
Performs a systematic search of the shortest communication pathways that traverse a protein structure. Wordom is a mixed Protein Structure Network (PSN) and Elastic Network Model (ENM)-based strategy, i.e., PSN-ENM, for fast investigation of allosterism in biological systems. The approach was validated on the PDZ2 domain from tyrosine phosphatase 1E (PTP1E) in its free (APO) and peptide-bound states.
Allows users to extract Protein Data Bank (PDB) information that bypasses the limitations associated with domain-based classifications. ConTemplate can suggest conformations for the query on the basis of its structural similarity to other proteins in the PDB, including proteins with low sequence similarity to the query. This tool can be applied on a genome-wide scale.
Produces quantitative scores to discriminate rheostat positions from those with “toggle” or “neutral” outcomes. RheoScale is a calculator of position scores, suitable for a wide range of experimental data in a wide range of proteins. It can also serve as a feature for protein positions in future prediction algorithms. The calculated positional scores enables predictions for personalized medicine and protein engineering. RheoScale is available as both a Microsoft Excel workbook and an R script.
A web server for detailed estimation of similarities between all pairs of 3D patterns detected in any two given protein structures. Geomfinder is composed of four main steps: (i) generates on each protein a virtual grid of coordinates which represents the initial location to find the 3D patterns; (ii) all possible 3D patterns occurring in each of the proteins of interest are detected, using as reference the virtual grid already generated; (iii) generates a list of four descriptors for each 3D pattern identified; (iv) makes use of the descriptors previously estimated to calculate a single similarity score among any possible pair of 3D patterns found in each queried protein.
Enables classification of new crystal structures. KinConform is a comprehensive machine learning based on classification of protein kinase active/inactive conformations. It takes any number of input structures, separates their chains and generates a fasta file of sequences. This fasta file is aligned (using MAPGAPS) to kinase profiles, identifying which chains are kinases. A series of measurements are then taken for each kinase chain and used as input to a machine learning classifier.
Performs protein structure predictions. AWSEM-MD is a module that extends the LAMMPS simulation platform with the aim of providing an alternative to structure-based models. The application couples local sequence similarity with a coarse-grained water mediated interaction. It also can be used for protein folding kinetics investigation as well as for studying functional dynamics of the native state, binding and folding processes.
Provides a refined version of a segment overlap score (SOV) tool based on the precedent definitions of SOV scores for evaluating the quality of protein secondary structure predictions. SOV_refine uses assignment of allowance for the overlapping segment pairs and scores for measuring similarity of different definitions of topologically associating domains (TADs). This software can handle unlimited number of states in a reference sequence.
Finds a maximum clique in an undirected graph. Maximum Clique Algorithm consists in an improvement to an approximate coloring algorithm. The improvement of this algorithm and the use of upper bounds, reduce the number of steps required to find the maximum clique. The software has been tested on random graphs and benchmark graphs, which were developed as part of the Second DIMACS Challenge.
Permits to obtain accurate model quality information for either individual or multiple uploaded models at the different levels of accuracy. SphereGrinder allows to evaluate structures from different points of view, to observe prediction difficulty and to identify promising predictions even if they seem to be inappropriate. It offers a solution for a comprehensive quality inspection conducted between a set of predicted protein models and a native structure.
Allows users to perform comparison and analysis for protein structure. PBE offers a platform composed of 10 tools which can accomplish various tasks such as (i) Encode protein structure into PBs sequence; (ii) Compare structure between a pair of proteins using PB description using both local and global alignment algorithms (iii) Mine a databank derived from SCOP, to detect proteins with similar fold (iv) Browse a database of preprocessed PB sequences and pairwise alignments at family and superfamily levels.
Sorts statistically significant profile-profile comparisons from the HHsearch output. HHsvm creates profiles of protein domains classified in SCOP database and employs HHsearch to compare each of these profiles with the entire database. It is suitable for automatic parsing of the large number of profile outputs. This tool is based on a support vector machine (SVM) method.
Allows user to visualize protein sequence features. GPCRHMM offers several advantages: users can zoom in on a region and zoom-dependent sequence rendering displays residues automatically at a sufficiently high zoom level. Moreover, researchers have control to customize the appearance of the data being displayed.