1 .TH "hmmsearch" 1 @RELEASEDATE@ "HMMER @RELEASE@" "HMMER Manual"
5 hmmsearch - search a sequence database with a profile HMM
20 for significantly similar sequence matches.
24 will be looked for first in the current working directory,
25 then in a directory named by the environment variable
27 This lets users use existing BLAST databases, if BLAST
28 has been configured for the site.
32 may take minutes or even hours to run, depending
33 on the size of the sequence database. It is a good
34 idea to redirect the output to a file.
37 The output consists of four sections: a ranked list
38 of the best scoring sequences, a ranked list of the
39 best scoring domains, alignments for all the best scoring
40 domains, and a histogram of the scores.
41 A sequence score may be higher than a domain score for
42 the same sequence if there is more than one domain in the sequence;
43 the sequence score takes into account all the domains.
44 All sequences scoring above the
48 cutoffs are shown in the first list, then
50 domain found in this list is
51 shown in the second list of domain hits.
52 If desired, E-value and bit score thresholds may also be applied
53 to the domain list using the
63 Print brief help; includes version number and summary of
64 all options, including expert options.
68 Limits the alignment output to the
72 shuts off the alignment output and can be used to reduce
73 the size of output files.
77 Set the E-value cutoff for the per-sequence ranked hit list to
81 is a positive real number. The default is 10.0. Hits with E-values
82 better than (less than) this threshold will be shown.
86 Set the bit score cutoff for the per-sequence ranked hit list to
91 The default is negative infinity; by default, the threshold
92 is controlled by E-value and not by bit score.
93 Hits with bit scores better than (greater than) this threshold
98 Calculate the E-value scores as if we had seen a sequence database of
100 sequences. The default is the number of sequences seen in your
108 Use the output format of HMMER 2.1.1, the 1998-2001 public
109 release; provided so 2.1.1 parsers don't have to be rewritten.
113 Sets the maximum number of CPUs that the program
114 will run on. The default is to use all CPUs
115 in the machine. Overrides the HMMER_NCPU
116 environment variable. Only affects threaded
117 versions of HMMER (the default on most systems).
121 Use Pfam GA (gathering threshold) score cutoffs.
123 to --globT <GA1> --domT <GA2>, but the GA1 and GA2 cutoffs
124 are read from the HMM file. hmmbuild puts these cutoffs there
125 if the alignment file was annotated in a Pfam-friendly
126 alignment format (extended SELEX or Stockholm format) and
127 the optional GA annotation line was present. If these
128 cutoffs are not set in the HMM file,
134 Use Pfam TC (trusted cutoff) score cutoffs. Equivalent
135 to --globT <TC1> --domT <TC2>, but the TC1 and TC2 cutoffs
136 are read from the HMM file. hmmbuild puts these cutoffs there
137 if the alignment file was annotated in a Pfam-friendly
138 alignment format (extended SELEX or Stockholm format) and
139 the optional TC annotation line was present. If these
140 cutoffs are not set in the HMM file,
146 Use Pfam NC (noise cutoff) score cutoffs. Equivalent
147 to --globT <NC1> --domT <NC2>, but the NC1 and NC2 cutoffs
148 are read from the HMM file. hmmbuild puts these cutoffs there
149 if the alignment file was annotated in a Pfam-friendly
150 alignment format (extended SELEX or Stockholm format) and
151 the optional NC annotation line was present. If these
152 cutoffs are not set in the HMM file,
158 Set the E-value cutoff for the per-domain ranked hit list to
162 is a positive real number.
163 The default is infinity; by default, all domains in the sequences
164 that passed the first threshold will be reported in the second list,
165 so that the number of domains reported in the per-sequence list is
166 consistent with the number that appear in the per-domain list.
170 Set the bit score cutoff for the per-domain ranked hit list to
174 is a real number. The default is negative infinity;
175 by default, all domains in the sequences
176 that passed the first threshold will be reported in the second list,
177 so that the number of domains reported in the per-sequence list is
178 consistent with the number that appear in the per-domain list.
180 only one domain in a sequence is absolutely controlled by this
183 The second and subsequent domains in a sequence have a de facto
184 bit score threshold of 0 because of the details of how HMMER
185 works. HMMER requires at least one pass through the main model
186 per sequence; to do more than one pass (more than one domain)
187 the multidomain alignment must have a better score than the
188 single domain alignment, and hence the extra domains must contribute
189 positive score. See the Users' Guide for more detail.
193 Use the Forward algorithm instead of the Viterbi algorithm
194 to determine the per-sequence scores. Per-domain scores are
195 still determined by the Viterbi algorithm. Some have argued that
196 Forward is a more sensitive algorithm for detecting remote
197 sequence homologues; my experiments with HMMER have not
198 confirmed this, however.
201 .BI --informat " <s>"
202 Assert that the input
206 do not run Babelfish format autodection. This increases
207 the reliability of the program somewhat, because
208 the Babelfish can make mistakes; particularly
209 recommended for unattended, high-throughput runs
210 of HMMER. Valid format strings include FASTA,
211 GENBANK, EMBL, GCG, PIR, STOCKHOLM, SELEX, MSF,
212 CLUSTAL, and PHYLIP. See the User's Guide for a complete
217 Turn off the post hoc second null model. By default, each alignment
218 is rescored by a postprocessing step that takes into account possible
219 biased composition in either the HMM or the target sequence.
220 This is almost essential in database searches, especially with
221 local alignment models. There is a very small chance that this
222 postprocessing might remove real matches, and
225 may improve sensitivity at the expense of reducing
226 specificity by letting biased composition hits through.
230 Run on a Parallel Virtual Machine (PVM). The PVM must
231 already be running. The client program
233 must be installed on all the PVM nodes.
234 Optional PVM support must have been compiled into
239 Turn on XNU filtering of target protein sequences. Has no effect
240 on nucleic acid sequences. In trial experiments,
242 appears to perform less well than the default
243 post hoc null2 model.
250 Master man page, with full list of and guide to the individual man
254 A User guide and tutorial came with the distribution:
260 Finally, all documentation is also available online via WWW:
261 .B http://hmmer.wustl.edu/
265 This software and documentation is:
268 HMMER - Biological sequence analysis with profile HMMs
269 Copyright (C) 1992-1999 Washington University School of Medicine
272 This source code is distributed under the terms of the
273 GNU General Public License. See the files COPYING and LICENSE
276 See the file COPYING in your distribution for complete details.
280 HHMI/Dept. of Genetics
281 Washington Univ. School of Medicine
283 St Louis, MO 63110 USA
284 Phone: 1-314-362-7666
286 Email: eddy@genetics.wustl.edu