1makehmmerdb(1)                   HMMER Manual                   makehmmerdb(1)
2
3
4

NAME

6       makehmmerdb - build a HMMER binary database file from a sequence file
7
8
9

SYNOPSIS

11       makehmmerdb [options] <seqfile> <binaryfile>
12
13
14

DESCRIPTION

16       makehmmerdb  is  used to create a binary file from a DNA sequence file.
17       This binary file may be used as a target database for  the  DNA  search
18       tool   nhmmer.Usingdefaultsettingsin  nhmmer,  this  yields  a  roughly
19       10-fold acceleration with small  loss  of  sensitivity  on  benchmarks.
20       (This  method  has been extensively tested, but should still be treated
21       as somewhat experimental.)
22
23
24

OPTIONS

26       -h     Help; print a brief reminder  of  command  line  usage  and  all
27              available options.
28
29
30
31
32
33

OTHER OPTIONS

35       --informat <s>
36              Assert  that  the  sequence  database  file  is  in  format <s>.
37              Accepted formats include fasta, embl,  genbank,  ddbj,  uniprot,
38              stockholm, pfam, a2m, and afa.  The default is to autodetect the
39              format of the file.
40
41
42
43       --bin_length <n>
44              Bin length. The binary file depends on a data  structure  called
45              the FM index, which organizes a permuted copy of the sequence in
46              bins of length <n>.  Longer bin  length  will  lead  to  smaller
47              files  (because  data  is  captured about each bin) and possibly
48              slower query time. The default is 256. Much more  than  512  may
49              lead to notable reduction in speed.
50
51
52
53       --sa_freq <n>
54              Suffix  array  sample  rate. The FM index structure also samples
55              from the underlying suffix array for the sequence database. More
56              frequent sampling (smaller value for <n>) will yield larger file
57              size and faster search (until file size becomes large enough  to
58              cause I/O to be a bottleneck). The default value is 8. Must be a
59              power of 2.
60
61
62
63       --block_size <n>
64              The input sequence is broken into blocks  of  size  <n>  million
65              letters. An FM index is built for each block, rather than buildā€
66              ing an FM index for the entire sequence database. Default is 50.
67              Larger blocks do not seem to yield substantial speed increase.
68
69
70
71

SEE ALSO

73       See  hmmer(1)  for  a master man page with a list of all the individual
74       man pages for programs in the HMMER package.
75
76
77       For complete documentation, see the user  guide  that  came  with  your
78       HMMER distribution (Userguide.pdf); or see the HMMER web page ().
79
80
81
82
84       Copyright (C) 2015 Howard Hughes Medical Institute.
85       Freely distributed under the GNU General Public License (GPLv3).
86
87       For  additional  information  on  copyright and licensing, see the file
88       called COPYRIGHT in your HMMER source distribution, or  see  the  HMMER
89       web page ().
90
91
92

AUTHOR

94       Eddy/Rivas Laboratory
95       Janelia Farm Research Campus
96       19700 Helix Drive
97       Ashburn VA 20147 USA
98       http://eddylab.org
99
100
101
102
103
104
105HMMER 3.1b2                      February 2015                  makehmmerdb(1)
Impressum