PubMed Health⌕ Search

Biomedical subjects

G M Church

Publications and source records attributed to G M Church.

At least 37 records · Page 2Linked to original sources
PubMed2000

Fast assignment of protein structures to sequences using the intermediate sequence library PDB-ISL.

MOTIVATION: For large-scale structural assignment to sequences, as in computational structural genomics, a fast yet sensitive sequence search procedure is essential. A new approach using intermediate sequences was tested as a shortcut to iterative multiple sequence search methods such as PSI-BLAST. RESULTS: A library containing potential intermediate sequences for proteins of known structure (PDB-ISL) was constructed. The sequences in the library were collected from a large sequence database using the sequences of the domains of proteins of known structure as the query sequences and the program PSI-BLAST. Sequences of proteins of unknown structure can be matched to distantly related proteins of known structure by using pairwise sequence comparison methods to find homologues in PDB-ISL. Searches of PDB-ISL were calibrated, and the number of correct matches found at a given error rate was the same as that found by PSI-BLAST. The advantage of this library is that it uses pairwise sequence comparison methods, such as FASTA or BLAST2, and can, therefore, be searched easily and, in many cases, much more quickly than an iterative multiple sequence comparison method. The procedure is roughly 20 times faster than PSI-BLAST for small genomes and several hundred times for large genomes. AVAILABILITY: Sequences can be submitted to the PDB-ISL servers at http://stash.mrc-lmb.cam.ac.uk/PDB_ISL/ or http://cyrah.ebi.ac.uk:1111/Serv/PDB_ISL/ and can be downloaded from ftp://ftp.ebi.ac.uk/pub/contrib/jong/PDB_+ ++ISL/ CONTACT: sat@mrc-lmb.cam.ac.uk and jong@ebi.ac.uk

Peptide Library↗
PubMed2000

Systematic management and analysis of yeast gene expression data.

We report steps toward the systematic management, standardization, and analysis of functional genomics data. We developed the ExpressDB database for yeast RNA expression data and loaded it with approximately 17.5 million pieces of data reported by 11 studies with three different kinds of high-throughput RNA assays. A web-based tool supports queries across the data from these studies. We examined comparability of data by converting data from 9 studies (217 conditions) into mRNA relative abundance estimates (ERAs) and by clustering of conditions by ERAs. We report on generation of ERAs and condition clustering for non-microarray data (5 studies, 63 conditions) and describe initial attempts to generate microarray-based ERAs (4 studies, 154 conditions), which exhibit increased error, on our web site http://arep.med.harvard. edu/ExpressDB. We recommend standards for data reporting, suggest research into improving comparability of microarray data through quantifying and standardizing control condition RNA populations, and also suggest research into the calibration of different RNA assays. We introduce a model for a database that integrates different kinds of functional genomics data, Biomolecule Interaction, Growth and Expression Database (BIGED).

Computational Biology↗
PubMed2000

Conservation of DNA regulatory motifs and discovery of new motifs in microbial genomes.

Regulatory motifs can be found by local multiple alignment of upstream regions from coregulated sets of genes, or regulons. We searched for regulatory motifs using the program AlignACE together with a set of filters that helped us choose the motifs most likely to be biologically relevant in 17 complete microbial genomes. We searched the upstream regions of potentially coregulated genes grouped by three methods: (1) genes that make up functional pathways; (2) genes homologous to regulons from a well-studied species (Escherichia coli); and (3) groups of genes derived from conserved operons. This last group is based on the observation that genes making up homologous regulons in different species are often assorted into coregulated operons in different combinations. This allows partial reconstruction of regulons by looking at operon structure across several species. Unlike other methods for predicting regulons, this method does not depend on the availability of experimental data other than the genome sequence and the locations of genes. New, statistically significant motifs were found in the genome sequence of each organism using each grouping method. The most significant new motif was found upstream of genes in the methane-metabolism functional group in Methanobacterium thermoautotrophicum. We found that at least 27% of the known E. coli DNA-regulatory motifs are conserved in one or more distantly related eubacteria. We also observed significant motifs that differed from the E. coli motif in other organisms upstream of sets of genes homologous to known E. coli regulons, including Crp, LexA, and ArcA in Bacillus subtilis; four anaerobic regulons in Archaeoglobus fulgidus (NarL, NarP, Fnr, and ModE); and the PhoB, PurR, RpoH, and FhlA regulons in other archaebacterial species. We also used motif conservation to aid in finding new motifs by grouping upstream regions from closely related bacteria, thus increasing the number of instances of the motif in the sequence to be aligned. For example, by grouping upstream sequences from three archaebacterial species, we found a conserved motif that may regulate ferrous ion transport that was not found in individual genomes. Discovery of conserved motifs becomes easier as the number of closely related genome sequences increases.

Computational Biology↗
PubMed2000

Regulatory networks revealed by transcriptional profiling of damaged Saccharomyces cerevisiae cells: Rpn4 links base excision repair with proteasomes.

Exposure to carcinogenic alkylating agents, oxidizing agents, and ionizing radiation modulates transcript levels for over one third of Saccharomyces cerevisiae's 6,200 genes. Computational analysis delineates groups of coregulated genes whose upstream regions bear known and novel regulatory sequence motifs. One group of coregulated genes contain a number of DNA excision repair genes (including the MAG1 3-methyladenine DNA glycosylase gene) and a large selection of protein degradation genes. Moreover, transcription of these genes is modulated by the proteasome-associated protein Rpn4, most likely via its binding to MAG1 upstream repressor sequence 2-like elements, that turn out to be almost identical to the recently identified proteasome-associated control element (G. Mannhaupt, R. Schnall, V. Karpov, I. Vetter, and H. Feldmann, FEBS Lett. 450:27-34, 1999). We have identified a large number of genes whose transcription is influenced by Rpn4p.

Algorithms↗
PubMed2000

Biclustering of expression data.

An efficient node-deletion algorithm is introduced to find submatrices in expression data that have low mean squared residue scores and it is shown to perform well in finding co-regulation patterns in yeast and human. This introduces "biclustering", or simultaneous clustering of both genes and conditions, to knowledge discovery from expression data. This approach overcomes some problems associated with traditional clustering methods, by allowing automatic discovery of similarity based on a subset of attributes, simultaneous clustering of genes and conditions, and overlapped grouping that provides a better representation for genes with multiple functions or regulated by many factors.

Algorithms↗
PubMed1999

In situ localized amplification and contact replication of many individual DNA molecules.

We describe a method to clone and amplify DNA by performing the polymerase chain reaction (PCR) in a thin polyacrylamide film poured on a glass microscope slide. The polyacrylamide matrix retards the diffusion of the linear DNA molecules so that the amplification products remain localized near their respective templates. At the end of the reaction, a number of PCR colonies, or 'polonies', have formed, each one grown from a single template molecule. As many as 5 million clones can be amplified in parallel on a single slide. If an Acrydite modification is included at the 5' end of one of the primers, the amplified DNA will be covalently attached to the polyacrylamide matrix, allowing further enzymatic manipulations to be performed on all clones simultaneously. We describe techniques to make replicas of these polony slides, and high throughput sequencing protocols for this technology. Other applications are also discussed.

Acrylic Resins↗
PubMed1999

Alignment and structure prediction of divergent protein families: periplasmic and outer membrane proteins of bacterial efflux pumps.

Broad-specificity efflux pumps have been implicated in multidrug-resistant strains of Pseudomonas aeruginosa and other Gram-negative bacteria. Most Gram-negative pumps of clinical relevance have three components, an inner membrane transporter, an outer membrane channel protein, and a periplasmic protein, which together coordinate efflux from the cytoplasmic membrane across the outer membrane through an unknown mechanism. The periplasmic efflux proteins (PEPs) and outer membrane efflux proteins (OEPs) are not obviously related to proteins of known structure, and understanding the structure and function of these proteins has been hindered by the difficulty of obtaining reasonable multiple alignments. We present a general strategy for the alignment and structure prediction of protein families with low mutual sequence similarity using the PEP and OEP families as detailed examples. Gibbs sampling, hidden Markov models, and other analysis techniques were used to locate motifs, generate multiple alignments, and assign PEP or OEP function to hypothetical proteins in several species. We also developed an automated procedure which combines multiple alignments with structure prediction algorithms in order to identify conserved structural features in protein families. This process was used to identify a probable alpha-helical hairpin in the PEP family and was applied to the detection of transmembrane beta-strands in OEPs. We also show that all OEPs contain a large tandem duplication, and demonstrate that the OEP family is unlikely to adopt a porin fold, in contrast to previous predictions.

Amino Acid Sequence↗
PubMed1999

Systematic determination of genetic network architecture.

Technologies to measure whole-genome mRNA abundances and methods to organize and display such data are emerging as valuable tools for systems-level exploration of transcriptional regulatory networks. For instance, it has been shown that mRNA data from 118 genes, measured at several time points in the developing hindbrain of mice, can be hierarchically clustered into various patterns (or 'waves') whose members tend to participate in common processes. We have previously shown that hierarchical clustering can group together genes whose cis-regulatory elements are bound by the same proteins in vivo. Hierarchical clustering has also been used to organize genes into hierarchical dendograms on the basis of their expression across multiple growth conditions. The application of Fourier analysis to synchronized yeast mRNA expression data has identified cell-cycle periodic genes, many of which have expected cis-regulatory elements. Here we apply a systematic set of statistical algorithms, based on whole-genome mRNA data, partitional clustering and motif discovery, to identify transcriptional regulatory sub-networks in yeast-without any a priori knowledge of their structure or any assumptions about their dynamics. This approach uncovered new regulons (sets of co-regulated genes) and their putative cis-regulatory elements. We used statistical characterization of known regulons and motifs to derive criteria by which we infer the biological significance of newly discovered regulons and motifs. Our approach holds promise for the rapid elucidation of genetic network architecture in sequenced organisms in which little biology is known.

Animals↗
PubMed1999

Quantifying DNA-protein interactions by double-stranded DNA arrays.

We have created double-stranded oligonucleotide arrays to perform highly parallel investigations of DNA-protein interactions. Arrays of single-stranded DNA oligonucleotides, synthesized by a combination of photolithography and solid-state chemistry, have been used for a variety of applications, including large-scale mRNA expression monitoring, genotyping, and sequence-variation analysis. We converted a single-stranded to a double-stranded array by synthesizing a constant sequence at every position on an array and then annealing and enzymatically extending a complementary primer. The efficiency of second-strand synthesis was demonstrated by incorporation of fluorescently labeled dNTPs (2'-deoxyribonucleoside 5'-triphosphates) and by terminal transferase addition of a fluorescently labeled ddNTP. The accuracy of second-strand synthesis was demonstrated by digestion of the arrayed double-stranded DNA (dsDNA) on the array with sequence-specific restriction enzymes. We showed dam methylation of dsDNA arrays by digestion with DpnI, which cleaves when its recognition site is methylated. This digestion demonstrated that the dsDNA arrays can be further biochemically modified and that the DNA is accessible for interaction with DNA-binding proteins. This dsDNA array approach could be extended to explore the spectrum of sequence-specific protein binding sites in genomes.

Base Sequence↗
PubMed1999

Modeling gene expression with differential equations.

We propose a differential equation model for gene expression and provide two methods to construct the model from a set of temporal data. We model both transcription and translation by kinetic equations with feedback loops from translation products to transcription. Degradation of proteins and mRNAs is also incorporated. We study two methods to construct the model from experimental data: Minimum Weight Solutions to Linear Equations (MWSLE), which determines the regulation by solving under-determined linear equations, and Fourier Transform for Stable Systems (FTSS), which refines the model with cell cycle constraints. The results suggest that a minor set of temporal data may be sufficient to construct the model at the genome level. We also give a comprehensive discussion of other extended models: the RNA Model, the Protein Model, and the Time Delay Model.

Computational Biology↗
PubMed1998

A comprehensive library of DNA-binding site matrices for 55 proteins applied to the complete Escherichia coli K-12 genome.

A major mode of gene regulation occurs via the binding of specific proteins to specific DNA sequences. The availability of complete bacterial genome sequences offers an unprecedented opportunity to describe networks of such interactions by correlating existing experimental data with computational predictions. Of the 240 candidate Escherichia coli DNA-binding proteins, about 55 have DNA-binding sites identified by DNA footprinting. We used these sites to construct recognition matrices, which we used to search for additional binding sites in the E. coli genomic sequence. Many of these matrices show a strong preference for non-coding DNA. Discrepancies are identified between matrices derived from natural sites and those derived from SELEX (Systematic Evolution of Ligands by Exponential enrichment) experiments. We have constructed a database of these proteins and binding sites, called DPInteract (available at http://arep.med.harvard.edu/dpinteract).

Bacterial Proteins↗
PubMed1998

Quantitative whole-genome analysis of DNA-protein interactions by in vivo methylase protection in E. coli.

A global methylation-based technique was used to identify, display, and quantitate the in vivo occupancy of numerous protein-binding sites within the Escherichia coli genome. The protein occupancy profiles of these sites showed variation across different growth conditions and genetic backgrounds. Of the 25 sites identified in this study, 24 occurred within 5' noncoding regions. Protein occupancy at 13 of these sites was supported by independent biochemical and genetic evidence. Most of the remaining 12 sites fell upstream of genes with no previously known function. A multivariate statistical analysis was utilized to group such uncharacterized genes with well-characterized ones, providing insights into their function based on a common pattern of transcriptional regulation.

DNA Modification Methylases↗
PubMed1998

Finding DNA regulatory motifs within unaligned noncoding sequences clustered by whole-genome mRNA quantitation.

Whole-genome mRNA quantitation can be used to identify the genes that are most responsive to environmental or genotypic change. By searching for mutually similar DNA elements among the upstream non-coding DNA sequences of these genes, we can identify candidate regulatory motifs and corresponding candidate sets of coregulated genes. We have tested this strategy by applying it to three extensively studied regulatory systems in the yeast Saccharomyces cerevisiae: galactose response, heat shock, and mating type. Galactose-response data yielded the known binding site of Gal4, and six of nine genes known to be induced by galactose. Heat shock data yielded the cell-cycle activation motif, which is known to mediate cell-cycle dependent activation, and a set of genes coding for all four nucleosomal proteins. Mating type alpha and a data yielded all of the four relevant DNA motifs and most of the known a- and alpha-specific genes.

Cloning, Molecular↗
PubMed1997

Comparing the predicted and observed properties of proteins encoded in the genome of Escherichia coli K-12.

Mining the emerging abundance of microbial genome sequences for hypotheses is an exciting prospect of "functional genomics". At the forefront of this effort, we compared the predictions of the complete Escherichia coli genomic sequence with the observed gene products by assessing 381 proteins for their mature N-termini, in vivo abundances, isoelectric points, molecular masses, and cellular locations. Two-dimensional gel electrophoresis (2-DE) and Edman sequencing were combined to sequence Coomassie-stained 2-DE spots representing the abundant proteins of wild-type E. coli K-12 strains. Greater than 90% of the abundant proteins in the E. coli proteome lie in a small isoelectric point and molecular mass window of 4-7 and 10-100 kDa, respectively. We identified several highly abundant proteins, YjbJ, YjbP, YggX, HdeA, and AhpC, which would not have been predicted from the genomic sequence alone. Of the 223 uniquely identified loci, 60% of the encoded proteins are proteolytically processed. As previously reported, the initiator methionine was efficiently cleaved when the penultimate amino acid was serine or alanine. In contrast, when the penultimate amino acid was threonine, glycine, or proline, cleavage was variable, and valine did not signal cleavage. Although signal peptide cleavage sites tended to follow predicted rules, the length of the putative signal sequence was occassionally greater than the consensus. For proteins predicted to be in the cytoplasm or inner membrane, the N-terminal amino acids were highly constrained compared to proteins localized to the periplasm or outer membrane. Although cytoplasmic proteins follow the N-end rule for protein stability, proteins in the periplasm or outer membrane do not follow this rule; several have N-terminal amino acids predicted to destabilize the proteins. Surprisingly, 18% of the identified 2-DE spots represent isoforms in which protein products of the same gene have different observed pI and M(r), suggesting they are post-translationally processed. Although most of the predicted and observed values for isoelectric point and molecular mass show reasonable concordance, for several proteins the observed values significantly deviate from the expected values. Such discrepancies may represent either highly processed proteins or misinterpretations of the genomic sequence. Our data suggest that AhpC, CspC, and HdeA exist as covalent homomultimers, and that IcdA exists as at least three isoforms even under conditions in which covalent modification is not predicted. We enriched for proteins based on subcellular location and found several proteins in unexpected subcellular locations.

Amino Acid Sequence↗
PubMed1997

Multiplex sequencing of 1.5 Mb of the Mycobacterium leprae genome.

The nucleotide sequence of 1.5 Mb of genomic DNA from Mycobacterium leprae was determined using computer-assisted multiplex sequencing technology. This brings the 2.8-Mb M. leprae genome sequence to approximately 66% completion. The sequences, derived from 43 recombinant cosmids, contain 1046 putative protein-coding genes, 44 repetitive regions, 3 tRNAs, and 15 tRNAs. The gene density of one per 1.4 kb is slightly lower than that of Mycoplasma (1.2 kb). Of the protein coding genes, 44% have significant matches to genes with well-defined functions. Comparison of 1157 M. leprae and 1564 Mycobacterium tuberculosis proteins shows a complex mosaic of homologous genomic blocks with up to 22 adjacent proteins in conserved map order. Matches to known enzymatic, antigenic, membrane, cell wall, cell division, multidrug resistance, and virulence proteins suggest therapeutic and vaccine targets. Unusual features of the M. leprae genome include large polyketide synthase (pks) operons, inteins, and highly fragmented pseudogenes.

Amino Acid Sequence↗
PubMed1997

Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization.

We have developed a new system of chromosomal mutagenesis in order to study the functions of uncharacterized open reading frames (ORFs) in wild-type Escherichia coli. Because of the operon structure of this organism, traditional methods such as insertional mutagenesis run the risk of introducing polar effects on downstream genes or creating secondary mutations elsewhere in the genome. Our system uses crossover PCR to create in-frame, tagged deletions in chromosomal DNA. These deletions are placed in the E. coli chromosome by using plasmid pKO3, a gene replacement vector that contains a temperature-sensitive origin of replication and markers for positive and negative selection for chromosomal integration and excision. Using kanamycin resistance (Kn(r)) insertional alleles of the essential genes pepM and rpsB cloned into the replacement vector, we calibrated the system for the expected results when essential genes are deleted. Two poorly understood genes, hdeA and yjbJ, encoding highly abundant proteins were selected as targets for this approach. When the system was used to replace chromosomal hdeA with insertional alleles, we observed vastly different results that were dependent on the exact nature of the insertions. When a Kn(r) gene was inserted into hdeA at two different locations and orientations, both essential and nonessential phenotypes were seen. Using PCR-generated deletions, we were able to make in-frame deletion strains of both hdeA and yjbJ. The two genes proved to be nonessential in both rich and glucose-minimal media. In competition experiments using isogenic strains, the strain with the insertional allele of yjbJ showed growth rates different from those of the strain with the deletion allele of yjbJ. These results illustrate that in-frame, unmarked deletions are among the most reliable types of mutations available for wild-type E. coli. Because these strains are isogenic with the exception of their deleted ORFs, they may be used in competition with one another to reveal phenotypes not apparent when cultured singly.

Alleles↗