UniqueProt: creating representative protein sequence sets

By Sven Mika, Burkhard Rost. 2003

132 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Better prediction of functional effects for sequence variants (2015, neutral): “We retrieved all enzyme pairs with pairwise sequence identity >40% and HSSPvalues>0 [28-30].”
UniRef: comprehensive and non-redundant UniProt reference clusters (2007, neutral): “A number of algorithms, tools and databases have been developed for clustering protein sequences (Enright and Ouzounis, 2000; Hobohm et al., 1992; Li et al., 2001; Mika and Rost, 2003; Paccanaro et al., 2006; Petryszak et al., 2005; Pipenbacher et al., 2002).”
Natively unstructured regions in proteins identified from contact predictions (2007, methodology): “UniqueProt (Mika and Rost, 2003) generated sequence-unique subsets: the maximal similarity between any protein used for training and testing was an HSSP-value<10 (Rost, 1999; Sander and Schneider, 1991), e.g. <31% pairwise sequence identity for >250 aligned residues.”
SSpro/ACCpro 5: almost perfect prediction of protein secondary structure and relative solvent accessibility using profiles, machine learning and structural similarity (2014, methodology): “Redundancy of the selected protein chains is first reduced at the 25% sequence identity level using CD-HIT (Li and Godzik, 2006) and Blastclust (Altschul et al ., 1997), and then further reduced using an HSSP cutoff distance of 5 using UniqueProt (Mika and Rost, 2003) resulting in 5772 protein chain”
PISCES: recent improvements to a PDB sequence culling server (2005, methodology): “Other criteria than strict sequence identity are sometimes used (7).”
Accurate Prediction of Protein Disordered Regions by Mining Protein Structure Data (2005, methodology): “We then filtered out homologous protein chains using UniqueProt (Mika and Rost, 2003) with a threshold HSSP value of 10.”
Ultrafast clustering algorithms for metagenomic sequence analysis (2012, methodology): “In the past, many available clustering programs were used for clustering protein sequences such as ProtoMap [18], ProtoNet [19], RSDB [20], GeneRAGE [21], TribeMCL [22], ProClust [23], UniqueProt [24], OrthMCL [25], MC-UPGMA [26], Blastclust [27] and CD-HIT [28–31].”
Predict impact of single amino acid change upon protein structure (2012, background): “The second set (dubbed “ hval0 ” ) contained 3,767 chains; it resulted from filtering at HVAL>0 [2,3,13] (corresponding to ~20% maximal pairwise sequence identity for alignments over 250 residues).”
LocTree3 prediction of localization (2014, unknown): “Sequenceredundancy was reduced at HVAL ≤ 0 (21,22) through UniqueProt (23).”
LocTree2 predicts localization for all domains of life (2012, methodology): “UniqueProt was used to reduce redundancy between the data sets and within each data set at HVAL > 0 and BLAST2 EVAL 10−3 with the minimum alignment length of 35 residues.”
Sequence-similar, structure-dissimilar protein pairs in the PDB (2007, unknown): “(6) ‘‘Protein-DNA/RNA’’—a DNA-bound protein versus its apo form.”
A Unified Multitask Architecture for Predicting Local Protein Properties (2012, background): “Their HSSP threshold of zero [44,45] corresponds to 20% pairwise sequence identity.”
Predicting transmembrane beta-barrels in proteomes. (2004, methodology): “We then ®ltered these using an HVAL of 0 (equation 1) as a metric in a greedy clustering algorithm identical to the protocol of UniqueProt (46).”
Minimizing proteome redundancy in the UniProt Knowledgebase (2016, background): “These include SkipRedundant (10), Decrease Redundancy (11), Pisces (12), UniqueProt (13), CD-HIT (14), FSA-BLAST (15), BlastClust (ftp://ftp.ncbi.nih.gov/ blast/documents/blastclust.html), MinSet (16), BlastCuller (17), FastCluster (18), Leaf (6), UCLUST (19) and Compressive genomics (20).”
Improved Disorder Prediction by Combination of Orthogonal Approaches (2009, methodology): “We generated sequence-unique subsets through UniqueProt [63] ascertaining that the pairwise sequence similarity between any pair of proteins corresponded to HSSPvalues,10 [64,65] which translated to ,31% pairwise sequence identity for .”