1 - 50 of 118 results

PFP / Protein Functional Prediction

Provides a web server that predicts Gene Ontology (GO) terms from a list of query sequences. PFP manages a large prediction coverage by retrieving annotations widely including from weakly similar sequences. The software uses predict function method which allows to consider sequences with a lack of annotated homologs in the database, extract and infers functional information. The tool can be combined with Extended Similarity Group (ESG) in a single interface.

ESG / Extended Similarity Group

Provides an improving specificity by accumulating contribution of consistently predicted Gene Ontology (GO) terms in an iterative search. ESG predicts a GO term with a high score if it appears many times consistently in the multiple searches including the initial search and the second level searches. The method can be applied in several multiple-domains and aims to improve predictions about functions derived from different domains. The tool can be combined with PFP (Protein Functional Prediction) in the a single interface.

EFICAz / Enzyme Function Inference by a Combined Approach

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.

SIFTS / Structure integration with function taxonomy and sequence

Includes cross-references to other biological resources such as Pfam, SCOP, CATH, GO, InterPro and the NCBI taxonomy database. The Structure Integration with Function, Taxonomy and Sequences resource (SIFTS) is focused on standardization of taxonomy information in the PDB based on the NCBI taxonomy database, and on adding cross-references to UniProtKB for all the protein sequences in the PDB that are present in the UniProt database. It has two main components—the semi-automated process that identifies the correct and up-to-date UniProtKB cross-reference for protein chains in the PDB and the automated pipeline that generates residue-level correspondences between proteins in the PDB and the corresponding UniProtKB sequence.


Performs annotation of carbohydrate-active enzymes and allows prediction of the enzymatic activity of the proteins. Hotpep (Homology to Peptide Pattern) is a state-of-the-art method for automatic annotation and functional prediction. The software matches the short, conserved motifs to undescribed protein sequences to obtain a fast, sensitive and precise annotation of carbohydrate-active enzymes to families. It also provides a functional prediction of function directly from amino acid sequence.

Argot / Annotation Retrieval of Gene Ontology Terms

Allows to predict protein function. The Argot strategy treats the grouping of Gene Ontology (GO) terms by means of semantic similarity in order to infer protein function. Compared to the previous version, the 2.5 version allows users to realize the selection of the input data from sequence similarity searches performed against a clustered version of UniProt databank and a remodeling of the weights given to Pfam hits. It’s also possible with the application of taxonomic constraints to classify annotations that cannot be applied to proteins belonging to the species through investigation.

POSSUM / POsition-Specific Scoring matrix-based featUre generator for Machine learning

Generates a broad spectrum of Position-Specific Scoring Matrix (PSSM)-based numerical representation schemes for protein sequences. POSSUM implements a wide range of algorithms available in the literature, provides an easy-to-use interface, and allows users to derive and customize the descriptors. More than 20 types of PSSM profile-based feature descriptors can be generated. The software facilitates feature generation, analysis, training and benchmarking of machine-learning models and predictions.


Utilizes an adaptive diffusion mechanism to improve the performance of BirgRank. AptRank introduces an adaptive mechanism to the PageRank framework that computes an optimal set of weights for the first several steps of diffusion so as to maximize recovery of a subset of known function annotations. It outperforms the four existing state-of-the-art methods in almost all cases, and in particular, outperforms those methods that do not incorporate information about the functional hierarchy.

INGA / Interaction Network GO Annotator

A web server developed to predict protein function from a combination of three orthogonal approaches. Sequence similarity and domain architecture searches are combined with protein-protein interaction network data to derive consensus predictions for GO terms using functional enrichment. The INGA server can be queried both programmatically through RESTful services and through a web interface designed for usability. The latter provides output supporting the GO term predictions with the annotating sequences. The method has been evaluated by the CAFA assessors (2014) among the best predictors.


Allows prediction of protein-to-protein and phenotype-to-protein functional associations based on phylogenetic profiling. ProtPhylo achieves flexibility and state-of-the-art taxonomic and functional coverage by generating phylogenetic profiles. It concerns more than 9 million non-redundant protein sequences across over 2000 organisms and implements four independent orthology detection algorithms. In summary, this tool allows prediction of subcellular localization, protein domains, membrane spanning regions, and complementary evidence of protein-protein interactions (PPIs).

NegGOA / Negative GO Annotations

Selects negative examples of GO annotations using ontology structure. NegGOA takes advantage of the ontology structure, available annotations and potentiality of additional annotations of a protein to choose negative examples of the protein. We compare this tool with other negative examples selection algorithms and find that NegGOA produces much fewer false negatives than them. We incorporate the selected negative examples into an efficient function prediction model to predict the functions of proteins in Yeast, Human, Mouse and Fly. It also demonstrates improved accuracy than these comparing algorithms across various evaluation metrics. In addition, NegGOA is less suffered from incomplete annotations of proteins than these comparing methods.

UniProt-DAAC / UniProt Domain Architecture Alignment and Classification

A method for the automatic annotation of protein sequences in the UniProt Knowledgebase (UniProtKB) by comparing their domain architectures, classifying proteins based on the similarities and propagating functional annotation. The performance of UniProt-DAAC was measured through a cross-validation analysis using the Gene Ontology annotation of a sub-set of Uni-ProtKB/Swiss-Prot. The results indicate the effectiveness of this approach in detecting functional similarity with an average F-score: 0.85.

MATEPred / Multidrug And Toxin Extrusion proteins PREDiction

Identifies Multidrug And Toxin Extrusion (MATE) proteins. MATEPred is based on Position Specific Scoring Matrix (PSSM) and uses Support Vector Machine (SVM). It returns sequence number, score and decision of the model. The tool was used to scan the proteomes of Vibrio parahaemolyticus and Shigella boydii for the presence of MATE proteins. It is able to differentiate MATE sequences from non-MATE sequences on the basis of PSSM profile.

PUFAS / PUFs Annotation Server

A web server that predicts the possible function of an input sequence, based upon the predicted domains and protein fold. The aim of the PUFAS server is to help identify the likely unknown function of a protein from its sequence derived from next generation sequencing data. It uses a series of methods, including homology search, ORF prediction, secondary structure, domains, and fold prediction, to identify the protein's likely assigned possible function of the genes unknown function.

PSIONplus / Predictor from Sequence of ION channels plus BLAST

Predicts ion channels proteins and their types, and subtypes of the voltage-gated ion channels. Empirical results show that combination of results generated by SVM model with the alignment by BLAST that is implemented in PSIONplus leads to improved predictive performance for the prediction of ion channels and voltage-gated channel subtypes when compared to using just BLAST. Results on the benchmark datasets that are independent of the datasets used to design our predictor reveal that PSIONplus obtains relatively good predictive performance. Its accuracy is 85.4% for the prediction of ion channels, 68.3% for the prediction of ion channel types, and its average accuracy is 96.4% for the prediction of the four subtypes of the voltage-gated channels.

MLDA / Multi-label Linear Discriminant Analysis

Detects function prediction in yeast proteins. MLDA uses sequence data from GenBank and network data form BioGRID. This tool was developed to deal with multi-label classification problems and meanwhile preserve the powerful classification capability of classical Linear Discriminant Analysis (LDA). Biological network data was incorporated on this method in a natural and integral way, such that the prediction performance is improved by taking advantage of the information from multiple different data sources.

GASS-WEB / Genetic Active Site Search-WEB

Can use catalytic and binding sites templates to search similar sites in a protein. GASS-WEB is a free and a user-friendly web server created for searching similar active sites based on data from the PDB, CSA and NCBI-VAST. This resource could be an invaluable tool for assisting protein function prediction and active site annotation. GASS-WEB can be used under two different scenarios: (i) given a protein of interest, to match a set of specific active site templates; or (ii) given an active site template, looking for it in a database of protein structures.

GROOLS / Genomic Rule Object-Oriented Logic System

Assists biocurators in functional annotation of proteins. GROOLS is an expert system using paraconsistent logic that evaluates the completeness and the consistency of predicted functions through biological processes like metabolic pathways. It evaluates the completeness and the consistency of predicted functions through biological processes like metabolic pathways. It could be integrated in an annotation system (such as MicroScope) as a live annotation companion to guide biologists during the curation process of their genome.

ProFET / Protein Feature Engineering Toolkit

Allows easy, universal discovery of new target proteins, as well as understanding the features underlying different high-level protein functions. ProFET extracts hundreds of features covering the elementary biophysical and sequence derived attributes. Most features capture statistically informative patterns. In addition, different representations of sequences and the amino acids (AA) alphabet provide a compact, compressed set of features. The results from ProFET were incorporated in data analysis pipelines, implemented in python, and adapted for multi-genome scale analysis.


A structure-based method for biological function annotation of protein molecules. To use COFACTOR, user needs to provide a 3D-structural model of the protein of interest. COFACTOR will thread the structure through the BioLiP protein function database by local and global structure matches to identify functional sites and homologies. Functional insights, including ligand-binding site, gene-ontology terms, and enzyme classification, will be derived from the best functional homology template.