help.txt

Usage:

To check IGGPIPE version:

    make version

To run the entire pipeline on your own data, invoke 'make' as follows:

     make PARAMS=myFilename ALL 2>&1 | tee logFiles/makeLog.txt

Here, 'myFilename' is the name of the parameter file you want to use to run the
pipeline.  This file would be an edited copy of allParameters.template, such as
'allParameters.XY' if for example you were using genomes X and y and named your
edited file with XY in the name.)  'ALL' is called the make 'target'.

There is a subdirectory named 'allParameters' containing a bunch of parameter
files I've tested.  You may want to put your own parameter files in here too, or
in a separate subdirectory you create.  If you do that, specify the full pathname
to 'make', e.g.:

     make PARAMS=allParameters/myFilename ALL 2>&1 | tee logFiles/makeLog.txt

Note that it may take many hours to run everything through to completion.  If an
error occurs, try to find the cause and fix it, then rerun the same 'make' command.
All steps that have already completed successfully are not re-run.

Also note that '2>&1 | tee logFiles/makeLog.txt' causes all 'make' standard and
error output to be written to text file 'makeLog.txt' in the 'logFiles' directory,
so you have a record of it, in addition to writing it to the console.  To view the
log you could use a command such as 'less logFiles/makeLog.txt'.  The '2>&1'
directs standard error output to the same place as regular output, and is optional.
We won't show the log file redirection below, but you can always use it.

Rather than running the entire pipeline all at once as shown above, you may
choose to run one pipeline step at a time and inspect its output before going
on to the next step.  To do this, run the following 'make' commands in the
order shown.  Each step specifies a different target name, and each one will
show the names of file(s) that are produced.

     make PARAMS=myFilename getSeqInfo              GENOME=ALL
     make PARAMS=myFilename getContigFile           GENOME=ALL
     make PARAMS=myFilename getKmers                GENOME=ALL
     make PARAMS=myFilename kmerStats               GENOME=ALL
     make PARAMS=myFilename kmersToText             GENOME=ALL
     make PARAMS=myFilename getGenomicPosIsect      GENOME=ALL
     make PARAMS=myFilename mergeKmers              GENOME=ALL
     make PARAMS=myFilename sortCommonUniqueKmers
     make PARAMS=myFilename findLCRs
     make PARAMS=myFilename findIndelGroups
     make PARAMS=myFilename getDNAseqsForPrimers    GENOME=ALL
     make PARAMS=myFilename findPrimers
     make PARAMS=myFilename ePCRtesting             GENOME=ALL
     make PARAMS=myFilename removeBadMarkers
     make PARAMS=myFilename plotMarkers

Several steps must be repeated for EACH GENOME.  In those steps, the genome
number is specified using GENOME=# on the command line, with # varying from 1
to N_GENOMES.  Or, use GENOME=ALL as shown above to automatically repeat the
step for all genomes at one time.  You could start with GENOME=ALL, but if you
have trouble, try doing it with GENOME=1 first.

You may want to run the first several steps, then check the number of common
unique k-mers to make sure you have plenty (say 8 million or more) but not too
many (say 40 million or more). You can run those steps and check the common
unique k-mer count (with genome 1 positions) with:

     make PARAMS=myFilename getGenomicPosIsect GENOME=1
     wc -l myOutFolder/Kmers/Kmers_1.isect

The 'wc' command shows a line count of file 'Kmers_1.isect', which contains one
unique k-mer per line.  If you have too few k-mers you should increase K by 1
and try again, and if you have too many, decrease it by 1 and try again.

You might then choose to run several more steps and then check the number of
LCRs obtained, using these commands:

     make PARAMS=myFilename findLCRs
     wc -l myOutFolder/LCR_*

The 'LCR_*' file contains LCRs, one per line.  If you had too few common k-mers
you might also have too few LCRs.  A million or more would be nice.  The fewer
you have, the fewer markers you are likely to get.

You might then choose to run the pipeline through to completion, then check the
number of non-overlapping markers obtained, using these commands:

     make PARAMS=myFilename ALL
     wc -l myOutFolder/MarkersNonoverlapping_*

Take a look at the plots in the .pdf and .png files.

For unexpected errors you can't fix, I may be able to help: twtoal@ucdavis.edu

If a step has already finished running successfully and you try to re-run that
step, 'make' will report that those files are up to date, and will do nothing.
To force 'make' to re-run a step, you must either delete the files created by
that step, or use the CLEAN=1 argument to tell 'make' to delete the files.  An
example of this is:

     make PARAMS=myFilename CLEAN=1 findLCRs

This deletes files created by the 'findLCRs' step.  If 'make' is re-run, with
the 'findLCRs' or 'ALL' target for example, it will rebuild those files and all
other files that were derived from them.  To delete ALL files that were built
by the pipeline, use this command:

     make PARAMS=myFilename CLEAN=1 ALL

If you want to delete all files in the output directory, regardless of their
names and whether or not they were built using the current parameter settings,
use this command (the output directory that is cleaned is defined in the PARAMS
file as DIR_IGGPIPE_OUT):

     make PARAMS=myFilename CLEAN_OUT_DIR=1 ALL

If you want to analyze an LCR, Indel Group, or marker file for Indels and SNPs
(a parameter should have already been set in the allParameters file to choose
which one), then these commands analyze for Indels and SNPs and plot Indel
information.  The last one, plotIndels, will cause the first two to also be
done, so it is the only one you need to perform the entire analysis from
finding K-mers to plotting Indels:

     make PARAMS=myFilename getDNAseqsForIndelsSNPs
     make PARAMS=myFilename IndelsSNPs
     make PARAMS=myFilename plotIndels