Runs cross_match with shorter option names and less noise, and reports the mismatch
level as mismatches over aligned bases rather than over query length. -perbasescore
adds a column giving the score as a percentage of the query's score against itself,
and -cg discounts one transition per CpG site. With -blast it runs rmblastn
through rmblast.pl instead of cross_match.
Self-compares a fasta file and drops sequences that another sequence already covers,
either by divergence (-divergence) or by score relative to the self-score
(-comparative). It can also merge overlapping fragments. For DNA it calls
Xmatch.pl from the same directory.
Run either script with no arguments for the full option list.
Perl 5, core modules only. Both scripts expect these on the PATH:
| Tool | Comes with | Needed for |
|---|---|---|
cross_match |
phrap | everything except -blast |
rmblast.pl |
RepeatModeler, in util/ |
-blast |
rmblastn, makeblastdb |
rmblast | -blast |
rmblast.pl sits in RepeatModeler's util/ subdirectory, which is often not on the
PATH:
export PATH="/usr/local/RepeatModeler/util:$PATH"Each script checks for the tools its options need before writing anything, and exits naming whatever is missing.
Matrices/ holds the scoring matrices. The scripts look for it next to themselves, so
keep the two scripts and Matrices/ in one directory.
Matrices/crossmatch/ holds the cross_match-format matrices and Matrices/ncbi/nt/
the rmblastn versions of the same ones. Under -blast, Xmatch.pl swaps the
directory in a -matrix path, so a matrix you name on the command line has to exist
in both.
| cross_match | 1.090518 |
| RepeatModeler | 2.0.7 |
| rmblastn | 2.17.1+ (BLAST 2.17.0) |
| Perl | 5.34.0 |