The PRINTS Database of Protein Fingerprints:  A Novel Information Resource for Computational Molecular Biology

Abstract

PRINTS is a compendium of protein motif fingerprints derived from the OWL composite sequence database. Fingerprints are groups of motifs within sequence alignments whose conserved nature allows them to be used as signatures of family membership. Fingerprints inherently offer improved diagnostic reliability over single motif methods by virtue of the mutual context provided by motif neighbors. To date, 650 fingerprints have been constructed and stored in PRINTS, the size of which has doubled in the last 2 years. The current version, 14.0, encodes 3500 motifs, covering a range of globular and membrane proteins, modular polypeptides, and so on. The database is now accessible via the UCL Bioinformatics Server on http:@ www.biochem.ucl.ac.uk/bsm/dbbrowser/. We describe here progress with the database, its compilation and interrogation software, and its Web interface.

Structural clues in the sequences of the aquaporins

“…Using the internal similarity (Figure 1) as a further constraint, the core segments were more narrowly de®ned as those occurring in both halves of each sequence, resulting in the truncation of core segments by zero to eight residues (Table 1). These segments clearly de®ne the transmembrane helices of 21-24 residues in length and the functional loops B and E. They also encapsulate regions of the sequences considered to be typical of the family in the databases BLOCKS (Henikoff & Henikoff, 1991) and PRINTS (Attwood et al, 1997) (Table 1). Figure 2 shows the topology of AQP1, with the six trans-membrane helices and two functional loops projecting into the plane of the membrane.…”

“…Table 1. Core segments derived from the alignment of the aquaporin family shown with respect to the AQP1 residues and compared with segments used in the PRINTS (Attwood et al, 1997) and BLOCKS (Henikoff & Henikoff, 1991) a LB(N), N-terminal non-conserved part of loop B linking to H2; LB(C), C-terminal non-conserved part of loop B linking to H3; LE(N), N-terminal non-conserved part of loop E linking to H5; LE(C), C-terminal non-conserved part of loop E linking to H6. The sum of hydrophobicity averages across the alignment for each segment compares with the minimum required for a transmembrane helix of $20 kcal/mol (Lee & Manoil, 1994).…”

Remote Thioredoxin Recognition Using Evolutionary Conservation and Structural Dynamics

“…Among the Prosite, PFAM, PANTHER (Mi et al, 2010), TIGRFAM (Haft et al, 2003), PRINTS (Attwood et al, 1997), CATH (Greene et al, 2007), and SCOP databases, there is significant overlap but no consensus, highlighting the difficulty in defining the boundaries (Petrey and Honig, 2009). In our study, we use the curated Prosite database as the gold standard.…”

Identification of host-microbe interaction factors in the genomes of soft rot-associated pathogens Dickeya dadantii 3937 and Pectobacterium carotovorum WPP14 with supervised machine learning

“…4) Finally, we incorporated over 20 basic gene or protein features (Table 3), such as GC content, amino acid composition and computed structural and physiochemical features of proteins and peptides [35], operon prediction [36], COG functional category [37], and codon adaptation index [38, 39]. Other gene features are derived from more complex analyses, including: (a) the phylogenetic profile method [40], which is based on the theoretical framework that co-occurrence of functionally linked proteins will be preserved by natural selection [41]; (b) Phylogenetic conservation which classifies genes according to distribution at different branching depths based on our phylogenetic framework for enterobacteria [11]; (c) PSORTb v3.0 [42] which predicts localization as cytoplasmic, cytoplasmic membrane, periplasmic, extracellular, or unknown; (d) Protein fingerprint scanning (a similarity search technique able to identify distantly related proteins) against identified fingerprints associated with virulence factors in PRINTS database [43, 44]; and (e) the gene neighbor method which identifies gene physical adjacency on a chromosome [45], based on the theory that neutral evolution tends to shuffle gene orders while functionally associated genes have conserved gene order. We employ both 150 bp and 300 bp as a threshold distance to define gene neighbors using ad hoc code.…”