Cannot run the basic flow in the homepage.

[EunCheon Lim]

Hi guys,

  I would like to choose a nice Multiple WGA algorithms for the 1001G datasets.

  I made a directory named alignathon. Then cloned mwgAlignAnalysis, mafTools, sonLib, and pinchesAndCacti in the folder. The binaries of mafTools are accessible from any scripts. I have executed "make all" command in each directory. The "make test" command yielded following errors:

  sonLib:
  ERROR: testFastaReadWriteC (bioioTest.TestCase)
  ...
  ERROR: testCigarReadWrite (cigarsTest.TestCase)
  ...
  ERROR: testSonLibKVTokyoCabinet (kvdbTest.TestCase)
  ...
  ERROR: testSonLibCTests (__main__.TestCase)

  pinchesAndCacti:
  ImportError: No module named cactus.shared.test

  ...

  Anyway, I ignored the failures in the tests.

  I downloaded a script, http://compbio.soe.ucsc.edu/alignathon/downloadTest.sh and it generated packageTest directory. Then, I runned

  make analysis location=...../alignathon/packageTest set=testSet

  This gives following errors:
  File "/ebio/abt6_projects9/ath_metagenome_20000_popindex/tmp/data/msa_test/alignathon/mwgAlignAnalysis/evaluations/bin/coverageCopyNumWrapper.py", line 47, in callEvaluation

  When I checked the script, I found that there is no mafCoveragePickleCreator.py, and mafCoveragePickleAnalysis.py files in the github repository.

  How could I run the sample package?

Regards,

Euncheon
=============================
PhD Student

Max Planck Institute for Developmental Biology
Department of Molecular Biology
Spemannstraße 37-39,
D-72076 Tuebingen, Germany

http://javawide.com

[EunCheon Lim]

In addition, mafTools are not compiling if I do not manually modify the common.mk file in the inc folder.
All flags with -Werror have to be changed to -Wno-error.

[Dent Earl]

Hi Euncheon!

Sorry for the incredible delay! Since the project ended four or so years ago we stopped looking at this group. I came across this thread while searching for something else. #1 I think the compilation things are now fixed, thanks for pointing them out. #2, regarding the missing mafCoveragePickle things, these were from an older version of mafTools. Here's some details of an email I wrote another interested researcher:

+benedict (as fyi, this is why we should publish the commit version of all our dependent libraries!)

I've been looking through the archives for these files and trying to read through the git commit history to discover where they may have been lost — 

These are from the mafTools (https://github.com/dentearl/mafTools) package. (This whole exercise has been a useful reminder that code can continue to live after a project ends and we should all publish the commit IDs of all of our dependencies so that future researchers can easily hook into exactly the same old code that was used).

It was removed in this commit:

https://github.com/dentearl/mafTools/commit/a47b47440186e9ebb38d2d00d165215dc7475193

It's still in the code at https://github.com/dentearl/mafTools/commit/4dc0a38d0761ab1e4d8702cee14422c55f299e72#diff-04c6e90faac2675aa89e2176d2eec7d8, so you could try to sync mafTools back to that commit. 

I think they were replaced by equivalent code in the mafCoverage and mafPairCoverage modules but I don't have the time at the moment to explore exactly how to refactor those two packages to hook into the old Alignathon code. Your best bet would be to sync your mafTools back to the old commit. Or just turn off that particular evaluation in the alignathon analysis code.

[Dent Earl]

+benedict for reals this time