TY - JOUR T1 - TIGRFAMs and Genome Properties in 2013. JF - Nucleic Acids Res Y1 - 2013 A1 - Haft, Daniel H A1 - Selengut, Jeremy D A1 - Richter, Roland A A1 - Harkins, Derek A1 - Basu, Malay K A1 - Beck, Erin KW - Databases, Protein KW - Genome, Archaeal KW - Genome, Bacterial KW - Genomics KW - Internet KW - Markov chains KW - Molecular Sequence Annotation KW - Proteins KW - sequence alignment AB -

TIGRFAMs, available online at http://www.jcvi.org/tigrfams is a database of protein family definitions. Each entry features a seed alignment of trusted representative sequences, a hidden Markov model (HMM) built from that alignment, cutoff scores that let automated annotation pipelines decide which proteins are members, and annotations for transfer onto member proteins. Most TIGRFAMs models are designated equivalog, meaning they assign a specific name to proteins conserved in function from a common ancestral sequence. Models describing more functionally heterogeneous families are designated subfamily or domain, and assign less specific but more widely applicable annotations. The Genome Properties database, available at http://www.jcvi.org/genome-properties, specifies how computed evidence, including TIGRFAMs HMM results, should be used to judge whether an enzymatic pathway, a protein complex or another type of molecular subsystem is encoded in a genome. TIGRFAMs and Genome Properties content are developed in concert because subsystems reconstruction for large numbers of genomes guides selection of seed alignment sequences and cutoff values during protein family construction. Both databases specialize heavily in bacterial and archaeal subsystems. At present, 4284 models appear in TIGRFAMs, while 628 systems are described by Genome Properties. Content derives both from subsystem discovery work and from biocuration of the scientific literature.

VL - 41 CP - Database issue M3 - 10.1093/nar/gks1234 ER - TY - JOUR T1 - TIGRFAMs and Genome Properties in 2013 JF - Nucleic acids researchNucleic Acids Research Y1 - 2013 A1 - Haft, Daniel H. A1 - J. Selengut A1 - Richter, Roland A. A1 - Harkins, Derek A1 - Basu, Malay K. A1 - Beck, Erin KW - Databases, Protein KW - Genome, Archaeal KW - Genome, Bacterial KW - Genomics KW - Internet KW - Markov chains KW - Molecular Sequence Annotation KW - Proteins KW - sequence alignment AB - TIGRFAMs, available online at http://www.jcvi.org/tigrfams is a database of protein family definitions. Each entry features a seed alignment of trusted representative sequences, a hidden Markov model (HMM) built from that alignment, cutoff scores that let automated annotation pipelines decide which proteins are members, and annotations for transfer onto member proteins. Most TIGRFAMs models are designated equivalog, meaning they assign a specific name to proteins conserved in function from a common ancestral sequence. Models describing more functionally heterogeneous families are designated subfamily or domain, and assign less specific but more widely applicable annotations. The Genome Properties database, available at http://www.jcvi.org/genome-properties, specifies how computed evidence, including TIGRFAMs HMM results, should be used to judge whether an enzymatic pathway, a protein complex or another type of molecular subsystem is encoded in a genome. TIGRFAMs and Genome Properties content are developed in concert because subsystems reconstruction for large numbers of genomes guides selection of seed alignment sequences and cutoff values during protein family construction. Both databases specialize heavily in bacterial and archaeal subsystems. At present, 4284 models appear in TIGRFAMs, while 628 systems are described by Genome Properties. Content derives both from subsystem discovery work and from biocuration of the scientific literature. VL - 41 N1 - http://www.ncbi.nlm.nih.gov/pubmed/23197656?dopt=Abstract ER - TY - JOUR T1 - Exopolysaccharide-associated protein sorting in environmental organisms: the PEP-CTERM/EpsH system. Application of a novel phylogenetic profiling heuristic JF - BMC biologyBMC biology Y1 - 2006 A1 - Haft, Daniel H. A1 - Paulsen, Ian T. A1 - Ward, Naomi A1 - J. Selengut KW - Amino Acid Motifs KW - Amino Acid Sequence KW - bacteria KW - Bacterial Proteins KW - Biofilms KW - Genome, Bacterial KW - Markov chains KW - Molecular Sequence Data KW - Phylogeny KW - Polysaccharides, Bacterial KW - Protein Sorting Signals KW - Protein Transport KW - Seawater KW - sequence alignment KW - Soil Microbiology AB - BACKGROUND: Protein translocation to the proper cellular destination may be guided by various classes of sorting signals recognizable in the primary sequence. Detection in some genomes, but not others, may reveal sorting system components by comparison of the phylogenetic profile of the class of sorting signal to that of various protein families. RESULTS: We describe a short C-terminal homology domain, sporadically distributed in bacteria, with several key characteristics of protein sorting signals. The domain includes a near-invariant motif Pro-Glu-Pro (PEP). This possible recognition or processing site is followed by a predicted transmembrane helix and a cluster rich in basic amino acids. We designate this domain PEP-CTERM. It tends to occur multiple times in a genome if it occurs at all, with a median count of eight instances; Verrucomicrobium spinosum has sixty-five. PEP-CTERM-containing proteins generally contain an N-terminal signal peptide and exhibit high diversity and little homology to known proteins. All bacteria with PEP-CTERM have both an outer membrane and exopolysaccharide (EPS) production genes. By a simple heuristic for screening phylogenetic profiles in the absence of pre-formed protein families, we discovered that a homolog of the membrane protein EpsH (exopolysaccharide locus protein H) occurs in a species when PEP-CTERM domains are found. The EpsH family contains invariant residues consistent with a transpeptidase function. Most PEP-CTERM proteins are encoded by single-gene operons preceded by large intergenic regions. In the Proteobacteria, most of these upstream regions share a DNA sequence, a probable cis-regulatory site that contains a sigma-54 binding motif. The phylogenetic profile for this DNA sequence exactly matches that of three proteins: a sigma-54-interacting response regulator (PrsR), a transmembrane histidine kinase (PrsK), and a TPR protein (PrsT). CONCLUSION: These findings are consistent with the hypothesis that PEP-CTERM and EpsH form a protein export sorting system, analogous to the LPXTG/sortase system of Gram-positive bacteria, and correlated to EPS expression. It occurs preferentially in bacteria from sediments, soils, and biofilms. The novel method that led to these findings, partial phylogenetic profiling, requires neither global sequence clustering nor arbitrary similarity cutoffs and appears to be a rapid, effective alternative to other profiling methods. VL - 4 N1 - http://www.ncbi.nlm.nih.gov/pubmed/16930487?dopt=Abstract ER - TY - JOUR T1 - A guild of 45 CRISPR-associated (Cas) protein families and multiple CRISPR/Cas subtypes exist in prokaryotic genomes JF - PLoS computational biologyPLOS Computational Biology Y1 - 2005 A1 - Haft, Daniel H. A1 - J. Selengut A1 - Mongodin, Emmanuel F. A1 - Nelson, Karen E. KW - Genes, Archaeal KW - Genes, Bacterial KW - Genes, Fungal KW - Genome KW - Genome, Bacterial KW - Haloarcula marismortui KW - Markov chains KW - Multigene Family KW - Oligonucleotide Array Sequence Analysis KW - Phylogeny KW - Prokaryotic Cells KW - Proteins KW - Repetitive Sequences, Nucleic Acid KW - Yersinia pestis AB - Clustered regularly interspaced short palindromic repeats (CRISPRs) are a family of DNA direct repeats found in many prokaryotic genomes. Repeats of 21-37 bp typically show weak dyad symmetry and are separated by regularly sized, nonrepetitive spacer sequences. Four CRISPR-associated (Cas) protein families, designated Cas1 to Cas4, are strictly associated with CRISPR elements and always occur near a repeat cluster. Some spacers originate from mobile genetic elements and are thought to confer "immunity" against the elements that harbor these sequences. In the present study, we have systematically investigated uncharacterized proteins encoded in the vicinity of these CRISPRs and found many additional protein families that are strictly associated with CRISPR loci across multiple prokaryotic species. Multiple sequence alignments and hidden Markov models have been built for 45 Cas protein families. These models identify family members with high sensitivity and selectivity and classify key regulators of development, DevR and DevS, in Myxococcus xanthus as Cas proteins. These identifications show that CRISPR/cas gene regions can be quite large, with up to 20 different, tandem-arranged cas genes next to a repeat cluster or filling the region between two repeat clusters. Distinctive subsets of the collection of Cas proteins recur in phylogenetically distant species and correlate with characteristic repeat periodicity. The analyses presented here support initial proposals of mobility of these units, along with the likelihood that loci of different subtypes interact with one another as well as with host cell defensive, replicative, and regulatory systems. It is evident from this analysis that CRISPR/cas loci are larger, more complex, and more heterogeneous than previously appreciated. VL - 1 N1 - http://www.ncbi.nlm.nih.gov/pubmed/16292354?dopt=Abstract ER - TY - Generic T1 - MCMC-based particle filtering for tracking a variable number of interacting targets. Y1 - 2005 A1 - Khan, Zia A1 - Balch, Tucker A1 - Dellaert, Frank KW - algorithms KW - Animals KW - Artificial Intelligence KW - Computer simulation KW - HUMANS KW - Image Enhancement KW - Image Interpretation, Computer-Assisted KW - Information Storage and Retrieval KW - Markov chains KW - Models, Biological KW - Models, Statistical KW - Monte Carlo Method KW - Motion KW - Movement KW - Pattern Recognition, Automated KW - Subtraction Technique KW - Video Recording AB -

We describe a particle filter that effectively deals with interacting targets--targets that are influenced by the proximity and/or behavior of other targets. The particle filter includes a Markov random field (MRF) motion prior that helps maintain the identity of targets throughout an interaction, significantly reducing tracker failures. We show that this MRF prior can be easily implemented by including an additional interaction factor in the importance weights of the particle filter. However, the computational requirements of the resulting multitarget filter render it unusable for large numbers of targets. Consequently, we replace the traditional importance sampling step in the particle filter with a novel Markov chain Monte Carlo (MCMC) sampling step to obtain a more efficient MCMC-based multitarget filter. We also show how to extend this MCMC-based filter to address a variable number of interacting targets. Finally, we present both qualitative and quantitative experimental results, demonstrating that the resulting particle filters deal efficiently and effectively with complicated target interactions.

JA - IEEE Trans Pattern Anal Mach Intell VL - 27 CP - 11 M3 - 10.1109/TPAMI.2005.223 ER - TY - JOUR T1 - The TIGRFAMs database of protein families JF - Nucleic acids researchNucleic Acids Research Y1 - 2003 A1 - Haft, Daniel H. A1 - J. Selengut A1 - White, Owen KW - Animals KW - Databases, Protein KW - Markov chains KW - Mixed Function Oxygenases KW - Phylogeny KW - Proteins KW - Pyruvate Carboxylase KW - Sequence Homology, Amino Acid AB - TIGRFAMs is a collection of manually curated protein families consisting of hidden Markov models (HMMs), multiple sequence alignments, commentary, Gene Ontology (GO) assignments, literature references and pointers to related TIGRFAMs, Pfam and InterPro models. These models are designed to support both automated and manually curated annotation of genomes. TIGRFAMs contains models of full-length proteins and shorter regions at the levels of superfamilies, subfamilies and equivalogs, where equivalogs are sets of homologous proteins conserved with respect to function since their last common ancestor. The scope of each model is set by raising or lowering cutoff scores and choosing members of the seed alignment to group proteins sharing specific function (equivalog) or more general properties. The overall goal is to provide information with maximum utility for the annotation process. TIGRFAMs is thus complementary to Pfam, whose models typically achieve broad coverage across distant homologs but end at the boundaries of conserved structural domains. The database currently contains over 1600 protein families. TIGRFAMs is available for searching or downloading at www.tigr.org/TIGRFAMs. VL - 31 N1 - http://www.ncbi.nlm.nih.gov/pubmed/12520025?dopt=Abstract ER -