Tolerating some redundancy significantly speeds up clustering of large protein databases

By Weizhong Li, Lukasz Jaroszewski, Adam Godzik. Bioinform., 2002

502 citations recorded by the source corpus.

Subfields: Molecular Biology, Spectroscopy

View Paper: https://www.semanticscholar.org/paper/a41302997f42a6ad5b8fab6050f422e5c67f0d2b

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

1. Yang Zhang
2. Johannes Söding
3. Burkhard Rost
4. Sitao Wu
5. Jinbo Xu
6. Andrej Sali
7. Weizhong Li
8. Andrei N. Lupas

In Spectroscopy

1. Burkhard Rost
2. Johannes Söding
3. Weizhong Li
4. Philip E. Bourne
5. Eric D. Scheeff
6. Adam Godzik
7. Andrei N. Lupas
8. Lukasz Jaroszewski

Citations in the corpus

• FastTree: Computing Large Minimum Evolution Trees with Profiles instead of a Distance Matrix (2009, methodology): “Also, before selecting sequences from the genuine alignment, we made a 99% non-redundant subset of sequences with CD-HIT ( Li et al., 2002 ).”
• WebMGA: a customizable web server for fast metagenomic sequence analysis (2011, methodology): “(cid:129) Sequence clustering has 4 tools: CD-HIT-EST, CD-HIT, H-CD-HIT [17-20] and CD-HIT-454 [31].”
• Sequence clustering strategies improve remote homology recognitions while reducing search times. (2002, methodology): “The web addresses are http://bioinformatics. burnham-inst.org/liwz/research/benchmark for the fold-recognition benchmark, http://bioinformatics.burnham-inst.org/cd-hi for the fast database clustering programs CD-HI and CD-HIT and http://bioinformatics.burnham-inst.org/pdb_blast for PDBBLAST search.”
• Predicting linear B-cell epitopes using string kernels (2008, unknown): “Because the resulting data set of 947 20-mer peptides was no longer non-redundant, we removed duplicated and highly homologous peptides by filtering the data set based on an 80% sequence identity cutoff using the CD-HIT program (Li et al., 2002) to obtain a homology-reduced data set of 701 peptides ”
• Clustered Sequence Representation for Fast Homology Search (2007, unknown): “Once a set of clusters have been identified, one sequence from each cluster is selected as the representative of that cluster to be inserted into the RSDB; the rest are discarded (Holm and Sander, 1998; Park et al., 2000; Li et al., 2001a, 2001b).”
• The Universal Protein Resource (UniProt) (2004, methodology): “UniRef90 and UniRef50 are built from UniRef100 using the CD-HIT algorithm (30) to provide non-redundant sequence collections for the scientific user community to perform faster homology searches.”
• Application of protein structure alignments to iterated hidden Markov model protocols for structure prediction (2006, methodology): “The nr databases used in this work were filtered down to a level of redundancy such that no sequence aligned to any other with greater than 60% identity (nr60), using the program cd-hit [47,48].”
• TPRpred: a tool for prediction of TPR-, PPR- and SEL1-like repeats from protein sequences (2007, methodology): “These profiles were generated by iterative searches against the non-redundant (NR) database at NCBI, filtered to a maximum pairwise sequence identity of 70% (NR-70) by CD-HIT [22,23].”
• Multiple Sequence Alignment Based on Profile Alignment of Intermediate Sequences (2007, methodology): “To avoid over-contribution of very similar intermediate sequences in the later profile construction step, we use CD-HIT (Li et al. 2002) to remove some of the closely related sequences so that the identity between the remaining intermediate sequences is less than 85%.”
• The Sorcerer II Global Ocean Sampling Expedition: Expanding the Universe of Protein Families (2007, methodology): “We clustered sequences in the MEROPS Peptidase Database [100] using CD-HIT [116,117] at 40% similarity level.”