Karplus, K., Kimmen Sj#olander, Barrett, C., Cline, M., Haussler, D., Hughey, R., Holm, L., & Sander, C. #1997#. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, Suppl. 1, 134#139.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is completely obscured in the first figure. 6 CONCLUSIONS AND FUTURE WORK This project is a continuation of earlier work developed to assist our computational biology group during the protein structure prediction contest CASP (Critical Assessment of Techniques for Protein Structure Prediction) [17] Our group used Leslie Grate s SAE, a prototype tool not intended for release) which combined an alignment editor with RasMol. After the contest, we decided that it would be useful both to create an alignment assessment tool for release and to develop additional ways to view structure sequence ....

....size, and strand width, thickness, or smoothness, It is somewhat cumbersome to rotate the entire molecule when the interest may lie in a small stretch of the protein. It may be useful to provide a pop up window for viewing a single amino acid This figure duplicated with author s permission. [17] substitution pair (as depicted in figure 7C) independently of the rest of the protein. We are also interested in viewing structure structure alignments (coordinate files for two protein structures that have been superimposed in three dimensions) Again, our streamline methods could be used to ....

Kevin Karplus, Kimmen Sjolander, Christian Barrett, Melissa Cline, and David Haussler. Predicting protein structure using hidden markov models. Proteins: Structure, Function, and Genetics, Supplement 1(1):134--139, 1997.

.... [4] is expected to complete in a few years, research focus has been shifted from sequencing the biological data to mining and interpreting these data [1, 7, 9] The interesting patterns to be mined range from genes [3] to DNA or protein sequence motifs [2, 10] to protein and RNA structure motifs [5, 9]. In this paper, we consider the problem of evaluating the significance of sequence motifs found by our pattern matching tool, Sdiscover [10] Given a set of sequences, the motifs of interest are in the regular expression form X 1 X 2 Delta Delta Delta, where each motif approximately matches ....

Karplus, K., Sjolander, K., Barrett, C., Cline, M., Haussler, D., Hughey, R., Holm, L., and Sander, C. Predicting protein structure using hidden Markov models. Proteins, Suppl. (1):134-- 139, 1997.

.... HMMs or related to HMMs, also for speech recognition, can be found in the collection of papers [24] Recently, HMMs have been applied to a variety of applications outside of speech recognition, such as handwriting recognition [25, 26, 27, 28, 29, 30, 31] pattern recognition in molecular biology [32, 33, 34, 35, 3], and fault detection [36] The variants and extensions of HMMs discussed here also include language models [37, 38, 12] econometrics [13, 14, 39] time series, and signal processing. The learning problem for the type of algorithms discussed here can be framed as follows. Given a training set D = ....

....them tractable) or make sure that the imperfections of the model do not hurt too much the final decision taking. These assumptions, however, have been found very useful in practice, in order to build the current state of the art speech recognition systems [16] and applications to bioinformatics [35, 3]. See also in [63] the use of a prior topology to discover sequential clusters in data. 3.4 Performance The performance of speech recognition systems based on HMMs varies a lot depending on the difficulty of the task. Benchmarks to compare speech recognition systems have been set up by ARPA [64] ....

K. Karplus, K. Sjolander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm, and C. Sander, "Predicting protein structure using hidden markov models," Proteins: Structure, Function and Genetics, vol. Supp. 1, no. 1, pp. 134--139, 1997.

No context found.

Karplus, K., Kimmen Sj#olander, Barrett, C., Cline, M., Haussler, D., Hughey, R., Holm, L., & Sander, C. #1997#. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, Suppl. 1, 134#139.

No context found.

Kevin Karplus, Kimmen Sjolander, Christian Barrett, Melissa Cline, David Haussler, Richard Hughey, Liisa Holm, and Chris Sander. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, Suppl. 1:134-139, 1997.

.... tools for sequence analysis [4] 5] 6] 7] Our efforts include participation in the international Human Genome Project [8] development of the gene finding software chosen to annotate the human and Drosophila (fly) genomes [9] success in international protein structure prediction contests [10], 11] and creation of a SIMD programmable accelerator for sequenceanalysis [12] Since the introduction of our first bioinformatics course in 1996, we have trained many students with majors in Computer Science, Computer Engineering, Mathematics, Molecular, Cell and Developmental (MCD) Biology, ....

Kevin Karplus, Kimmen Sjolander, Christian Barrett, Melissa Cline, David Haussler, Richard Hughey, Liisa Holm, and Chris Sander, "Predicting protein structure using hidden Markov models," Proteins: Structure, Function, and Genetics, vol. Suppl. 1, pp. 134--139, 1997.

....the initial basis, the associated HSSP alignment [22] of the structure and its homologs. This alignment and the corresponding hmm parameters were re estimated using standard hmm methods in combination with priors over amino acids and transition probabilities in various structural environments (see [14]) The transition priors allowed us to incorporate general structural information, such as the low probability of an insert in the middle of a helix. Following re estimation, we applied sequence weighting (Section 2.3) to generalize the models for recognition of remote homologs. 2.2 Building the ....

....This initial model was used to select homologs from a set of neighbors from the Entrez database. The model parameters were reestimated repeatedly on the target sequence and homologs, using Dirichlet mixture densities over amino acid distributions and a variety of different transition priors (see [14] for details) 3 Because proteins can have repeated domains, the multdomain module of SAM was used to select subsequences from the putative homolog set. For instance, some homologs to t0004 (the nucleotidyltransferase S1 motif) had 3 or 4 regions that matched the model. The alignment of the ....

[Article contains additional citation context not shown here]

K. Karplus, Kimmen Sjolander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm, and C. Sander. Predicting protein structure using hidden Markov models, the CASP2 contest. Technical Report UCSC-CRL-97-13, University of California, Santa Cruz, Computer Science, UC Santa Cruz, CA 95064, 1997.

....outside the family. The probability score can thus be interpreted as a measure of the extent to which a new protein sequence is homologous to the protein family of interest. Considerable recent work has been done in refining HMMs for the purpose of identifying weak protein homologies in this way [34, 5, 14, 24, 33]. Let X = x 1 , x n ] denote a protein sequence, where each x i is an amino acid residue. Suppose that we are interested in a particular protein family such as immunoglobulins and have estimated an HMM, H 1 , for this family (for details of the estimation process see, e.g. 12] We use P ....

....kernel function mediates all the pairwise comparisons between the protein sequences. Our approach here is to derive the kernel function from HMMs corresponding to the protein family of interest. We are thus able to build on the work of others towards adapting HMMs for protein homology detection[34, 24, 33]. Our use of protein models in the kernel function, however, deviates from the standard use of such models in biosequence analysis. More precisely, the kernel function specifies a similarity score for any pair of sequences, whereas the likelihood score from an HMM only measures the closeness of ....

Kevin Karplus, Kimmen Sjolander, Christian Barrett, Melissa Cline, David Haussler, Richard Hughey, Liisa Holm, and Chris Sander. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, Suppl. 1:134-- 139, 1997.

....the associated HSSP alignment [22] of the structure and its homologs as the initial basis. This alignment and the corresponding hmm parameters were re estimated using standard hmm methods in combination with priors over amino acids and transition probabilities in various structural environments [14]. The transition priors allowed us to incorporate general structural information, such as the low probability of an insert in the middle of a helix. Following re estimation, we applied sequence weighting (Section 2.3) to generalize the models for recognition of remote homologs. 2.2 Building the ....

....This initial model was used to select homologs from a set of neighbors from the Entrez database. The model parameters were reestimated repeatedly on the target sequence and homologs, using Dirichlet mixture densities over amino acid distributions and a variety of different transition priors [14]. Because proteins can have repeated domains, the multdomain module of SAM was used to select subsequences from the putative homolog set. For instance, some homologs to t0004 (the nucleotidyltransferase S1 motif) had 3 or 4 regions that matched the model. The alignment of the target and homologs ....

[Article contains additional citation context not shown here]

K. Karplus, Kimmen Sj¨olander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm, and C. Sander. Predicting protein structure using hidden Markov models, the CASP2 contest. Technical Report UCSC-CRL-97-13, University of California, Santa Cruz, Computer Science, UC Santa Cruz, CA 95064, 1997.

....application of the combined classifier, we consider the well known problem of recognizing remote homologies (evolutionary structural similarities) between protein sequences 8 that have low residue identity. Considerable recent work has been done in refining hidden Markov models for this purpose [7, 2, 3, 6], and such models current achieve the best performance 9 . We use these stateof the art HMMs as comparison cases and also as sources for deriving the kernel function. Here we used logistic regression with the simple kernel KU (X i ; X j ) as the number of parameters in the HMMs was several ....

K. Karplus, Kimmen Sj¨olander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm, and C. Sander. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, Supplement 1(1):134-- 139, 1997.

....al. 1993; Krogh et al. 1994) or generalized profiles (Bucher Bairoch, 1994) have been demonstrated to be very effective in detecting conserved patterns in multiple sequences (Hughey Krogh, 1996; Baldi et al. 1994; Eddy et al. 1995; Eddy, 1995; Bucher et al. 1996; McClure et al. 1996; Karplus et al. 1997; Grundy et al. 1997; Karchin Hughey, 1998) The typical profile hidden Markov model (Figure 1) is a chain of match (square) insert (diamond) and delete (circle) nodes, with all transitions between nodes and all character costs in the insert and match nodes trained to specific probabilities. ....

....the reversed sequence, as does the lower frequency surface core hydrophobicity pattern. Because of these subtle effects, the reversed sequence is a much more realistic decoy than a scrambled sequence. These effects can affect scoring signifcantly. For example, in the scoring for the CASP 2 contest (Karplus et al. 1997), we had to eliminate by hand some coiled coil models that scored any helical protein well the reversedmodel scoring eliminates these problems. Also, metallothionein (4mt2) with 24 cysteines out of 61 residues, can align well to almost any sequence with conserved cysteines. Since many hmms get ....

Karplus, K., Kimmen Sj¨olander, Barrett, C., Cline, M., Haussler, D., Hughey, R., Holm, L., & Sander, C. (1997). Predicting protein structure using hidden Markov models.

No context found.

Karplus, K., Sjlander, K., Barret, C., Cline, M., Haussler, D., Hughey, R., Holm, L. & Sander, C. (1997). "Predicting protein Structure using hidden Markov models". Proteins: Structure, Function, and Genetics. 134-139.

No context found.

) K. Karplus, K. Sjolander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm, and C. Sander. Predicting protein structure using hidden Markov models. Proteins: Structure, Function, and Genetics, pages 134--139, 1997. supplement 1.

No context found.

Karplus,K., Sjolander,K., Barrett,C., Cline,M., Haussler,D., Hughey,R., Holm,L. and Sander,C. (1997) Predicting protein structure using hidden Markov models. Proteins, 1(Suppl.), 134--139.

No context found.

Karplus,K., Sjlander,K., Barrett,C., Cline,M., Haussler,D., Hughey,R., Holm,L. and Sander, C. (1997) Predicting protein structure using hidden Markov models. Proteins Struct. Funct. Genet., Suppl. 1, 134--139.

No context found.

K. Karplus, K. Sjolander, C. Barrett, M. Cline, D. Haussler, R. Hughey, L. Holm and C. Sander, Predicting Protein Structure using Hidden Markov Models, Proteins: Structure, Function, and Genetics, Supplement , (1), pp. 134-139, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC