This is an automated email from the git hooks/post-receive script. sascha-guest pushed a commit to branch master in repository augustus.
commit 3d0e9ff1ead87458bbced9b7d902a89a04475983 Author: Sascha Steinbiss <[email protected]> Date: Thu Mar 31 17:45:59 2016 +0000 add manpages contributed by upstream --- debian/mansrc/fastBlockSearch.1.adoc | 50 ++++++++++++++++++++++++++++++++++++ debian/mansrc/prepareAlign.1.adoc | 34 ++++++++++++++++++++++++ 2 files changed, 84 insertions(+) diff --git a/debian/mansrc/fastBlockSearch.1.adoc b/debian/mansrc/fastBlockSearch.1.adoc new file mode 100644 index 0000000..9a00272 --- /dev/null +++ b/debian/mansrc/fastBlockSearch.1.adoc @@ -0,0 +1,50 @@ +# fastBlockSearch(1) + +## NAME + +fastBlockSearch - search loci matching protein block profiles + +## SYNOPSIS + +*fastBlockSearch* [options] seqs.fa fam.prfl + +## DESCRIPTION + +Searches hits (matches) of the blocks in the profile given by fam.prfl within +the genomic sequences in the file seqs.fa. +Hits are sorted increasingly by score, so the last displayed hit is the best one +found in the region. The format is similar to that of the blockscore file (which +is optionally generated by msa2prfl.pl): It shows coordinate, strand, mean odds- +ratio score, and specificity of score, and the motif. +From the output users can chose regions with matching blocks to perform gene +prediction with AUGUSTUS-PPX using the same block profile. + +## OPTIONS + +*--cutoff=c*:: + This minimum for the average log score of the motifs found can be used to adjust the sensitivity of the block search. + The standard cutoff is 0.7, which is very sensitive but can give many false positive hits for smaller profiles. + +## EXAMPLE + + > fastBlockSearch --cutoff=1.1 chr4.103M.fa PF00225_seed.prfl + + Hits found in chr4 103000000 105000000 + Score:207.987 + Mult. score:4.83391 + 1081586 unknown_M[5,13] - 2.32574 5.04633 .....YATRLKNI + 1103952 unknown_L - 4.85363 6.75245 NAKTRIICTITP + 1103991 unknown_K - 8.38065 9.92928 YRDSKLTRILQNSLG + 1104375 unknown_J - 3.96065 6.79408 RSLFILGQVIKKL + 1106992 unknown_I - 9.22487 7.64306 LVDLAGSE + 1115567 unknown_H[5,16] - 2.31869 5.58986 .....ESRHYGETKMN + 1116319 unknown_G - 7.34282 8.29425 EIYNETITDLL + 1117092 unknown_F - 5.10694 6.10274 VIPRAIHDIF + 1117146 unknown_E - 9.43596 9.18891 QTASGKTYTM + 1117176 unknown_D[1,8] - 5.73796 6.31532 .GTIFAYG + 1117399 unknown_B[1,7] - 3.59083 5.03059 .CLDRVF + 1119420 unknown_A[0,8] - 4.64107 6.44285 RVRPLNSR. + +## AUTHORS + +Oliver Keller diff --git a/debian/mansrc/prepareAlign.1.adoc b/debian/mansrc/prepareAlign.1.adoc new file mode 100644 index 0000000..818c5f4 --- /dev/null +++ b/debian/mansrc/prepareAlign.1.adoc @@ -0,0 +1,34 @@ +# prepareAlign(1) + +## NAME + +prepareAlign - prepare an alignment for block profile creation by deleting gap-rich rows + +## SYNOPSIS + +*prepareAlign* < input.fa > output.fa + +## DESCRIPTION + +An input multiple alignment of proteins is read from standard input. +prepareAlign removes rows from it with the aim of producing fully conserved +blocks in the remaining alignment which is then output to standard output. + +When an alignment contains many sequences and too many gaps, e.g. a full PFAM +alignment, trying to run msa2prfl.pl results in the message "No blocks found in +MSA". Then use "prepareAlign" first to eliminate sequences. + +## AUTHORS + +Oliver Keller + + + + + + + + + + + -- Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-med/augustus.git _______________________________________________ debian-med-commit mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-commit
