Hello,
I've generated a multiple sequence alignment pf protein sequences with HMMER suite. The possible output formats with this are SELEX and STOCKHOLM. Is there anyway to convert these formats to FASTA multiple sequence alignment format? I've found this but it is not accepting large files.
esl-* tools ship with HMMER, don't they?
esl-reformat -h
# esl-reformat :: convert between sequence file formats
# Easel h3.1b1 (May 2013)
# Copyright (C) 2013 Howard Hughes Medical Institute.
# Freely distributed under the Janelia Farm Software License.
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Usage: esl-reformat [-options] <format> <seqfile>
Output format choices: Unaligned Aligned
----------- -------
fasta a2m
hmmpgmd afa
clustal
clustallike
pfam
phylip
phylips
psiblast
selex
stockholm
where options are:
-d : convert to DNA alphabet (U->T)
-h : help; print brief info on version and usage
-l : convert to lower case
-n : remove DNA IUPAC codes; convert ambig chars to N
-o <s> : send output to file <f>, not stdout
-r : convert to RNA alphabet (T->U)
-u : convert to upper case
-x : convert non-IUPAC chars (e.g. X) in DNA to N
--gapsym <s> : convert all gaps to character <c>
--informat <s> : input sequence file is in format <s>
--mingap : remove columns containing all gaps (seqfile=MSA)
--keeprf : with --mingap, keep all nongap #=GC RF columns
--nogap : remove columns containing any gaps (seqfile=MSA)
--wussify : convert old RNA structure markup lines to WUSS
--dewuss : convert WUSS RNA structure markup to old format
--fullwuss : convert simple WUSS notation to full (output) WUSS
--ignore <s> : ignore input seq characters listed in string <s>
--acceptx <s> : accept input seq chars in string <s> as X
--rename <s> : rename and number each sequence <s>.<n>
--replace <s> : <s> = <s1>:<s2> replace characters in <s1> with those in <s2>
--small : use minimal RAM, input must be pfam, output must be afa or pfam
--id_map <s> : if format is hmmpgmd, put the id map into file <s>
Easel can be downloaded from here.
The instruction on how to build the source code is in the link.
The esl-reformat is in the miniapps folder.
./esl-reformat **a2m** seqfile > alignfile
If one is using hmmalign to create the alignments, it can output a2m format:
hmmalign -**-outformat A2M** -o alignment.a2m
This script or this script should do that. Dependent on BioPerl. Try them and validate the output. If does not work try to modify the script.
•
written
by
GouthamAtla
Log in to answer this question.
