Commits

Author Commit Message Labels Comments Date
James Taylor
Stuff, some support for maf i and e rows, a new tiling script.
David King
added some *more* annotation checks to gene_reader.FeatureReader\ and bitset_intersect and list2bits to bitset_utils.py
David King
added bitset_interval_intersect to utils
David King
gene_reader.py uses bitset_utils now
David King
bitset_utils.py: some cleanup
James Taylor
Another script I had in several places so it is probably useful enough to go in here.
David King
name change
David King
fourfold_degenerate_translate.py: bed-like output labels degenerate codon positions. Takes bed or gtf files.
David King
fixed indent errors-- sometimes I think I'm in C
David King
gene_reader.py: Readers for gtf files
David King
bitset_utils.py: basic utils for lists, taken semi-verbatim from bitset_builders that operate on files
James Taylor
Added a check in "_seqmapping.pyx" to ensure the input buffer contains 32bit elements. Numeric defaults to 64bit on 64bit platforms (apparently, it does on herbie at least) so this makes it a little hard to accidently screw up. Type checking is hard with the buffer protocol. Tests should pass on both platforms now. Oh, and some things in here now require python2.4
James Taylor
Updated pkg_resources.py
David King
move pwm to bx/pwm
James Taylor
Removed seq_numarray and alphabet.py, replaced with much better code taken from RP. The only thing actually using this was 'maf_to_int_seqs', but I think it is pretty damn useful and belongs here rather than RP. Only things still using numarray seem to be the pwm stuff and the hdf5 related stuff. The latter must use it since pytables does not support Numeric. Medium term I'd like to make everything here use the new 'numpy' once they cut the 1.0 release and eliminate any dependence on Numeric or numarray. Pytables and matplotlib will both be using numpy, and it fixes some of the speed problems with numarr…
James Taylor
Moved maf tests into the conventional place, all tests except test_seq_numarray appear to be good. I'm going to purge all the numarray crap shortly.
James Taylor
Okay, tests all look good.
James Taylor
Moved a bunch of stuff that we didn't write into the 'bx_extras' package. Removed the broken kdtree implementation. Fixed up some tests.
James Taylor
Phast tree model reader tests, pyrex optomization for alignments.
James Taylor
Tests work (except some pieces I haven't committed, will fix later)
James Taylor
Big partial commit, spring cleaning!
James Taylor
Fixing up phast reader, removing scipy dependency (need to abstract the linalg stuff somehow), added tests. Yay tests!
James Taylor
Very rudimentary support for PHAST's tree model file format.
James Taylor
Newick format tree parser using pyparsing (sweet!). Some refactoring and documentation in bed_rand_intersect.
James Taylor
Tool for extracting colum counts from a MAF.
James Taylor
Adding the stats package which is used by a few things. This is 3rd party code, and might cause namespace conflicts when used with other eggs also containing this module. Still need a better way to deal with that in general, but it should work okay for now (warnings maybe but not fatal)
James Taylor
Refactored bx.intervals.io to extend the stuff from bx.tabular.io Other minor changes to documentation and tests.
James Taylor
Another script for adding columns to a table. Also, the expression is optional for the filter so it can be used as a tabular cut as well. For example, suppose I have a table with many columns labeled X_Pos, Y_Pos, ... which contain 0 if some experiment was succesfull or 1 otherwise. I want to add a column with the sum of these experiments. So: cat combined_table.txt | ~/projects/bx-python-trunk/table_add_column.py \ -H "sum( int( row[x] ) for x in row.fieldnames if x.endswith( '_Pos' ) )" NPos
James Taylor
Added classes for working with tabular data files. The intervals stuff should probably be refactored to extend this (tables of intervals are tables). There is a new script called 'table_filter.py' for filtering and cutting tables from the command line. Allows you to easily do things like: """ cat combined_table.txt | table_filter.py -H -c Acc,5way_rp "row['5way_rp'] != 'NA' and row['5way_rp'] > 0" """ Which returns the Acc and 5way_rp columns from combined_table.txt where the 5way_rp value is both not NA and > 0, keeping the header.
James Taylor
Added a tool for joing two lists of intervals side by side. Fixes for interval io.
  1. Prev
  2. Next