quikr.git - Quikr - a bio classifier based on NNLS

Age	Commit message (Collapse)	Author
2013-03-14	get rid of OS checks for count, assume it is called count-kmers	Calvin

2013-03-14	re-add the kmer filename checK	Calvin

2013-03-13	fix bad old commit	Calvin

2013-03-13	added multi_fasta_otu to cli.markdown	Calvin

2013-03-13	added matlab markdown and edited cli	Calvin

2013-03-13	added setup.py for python quikr module	Calvin

2013-03-13	updated makefile to install nbc properly and python properly.	Calvin

2013-03-12	initial commit of some documentation	Calvin

2013-03-12	removed multifasta's py extension, modified the --help argument output for ↵	Calvin
	quikr_train
2013-03-11	removed documented matlab, add bare makefile	Calvin

2013-03-11	move python stuff to python directory	Calvin

2013-03-11	adding matlab code	Calvin

2013-03-08	moving some stuff	Calvin

2013-03-08	fix quikr_train to work with updated quikr module	mutantturkey

2013-03-08	convert into a module, rename out functions so all quikr functions are ↵	mutantturkey
	withing quikr.function name
2013-03-08	fix quikr	mutantturkey

2013-03-08	fix multifasta_to_otu	mutantturkey

2013-03-07	starting to modularize	Calvin

2013-03-07	minor formatting cleanup	Calvin

2013-03-07	multifasta_to_otu.py support for compression	Calvin

2013-03-07	added gzip compression support	Calvin

2013-03-07	Check if file is compressed	Calvin

2013-03-06	removed debugging statements	Calvin

2013-03-06	do the matrix rotations and manipulations in quikr_train instead of quikr	Calvin

2013-03-06	Cleanups and refactoring	Calvin
	* Only do one check for output_directory by combining statements * rename output to number_of_reads * get rid of useless comment
2013-03-06	fixed index	Calvin

2013-03-06	formatting cleanups, removed unused import, deleted unused lines	Calvin

2013-03-05	Added OTU table output and grep parsing	Calvin
	A giant batch commit. Sorry! * OTU table output * results are still different from matlab * sample names still have suffix * otherwise should be ok * only return rows which sum up to greater than zero (ie non empty) * Use grep instead of BIOseq because it is a lot faster to count our sequences this way. * re-enable our pool process management * refactor and rename records to headers * refactor final_output to output
2013-03-05	simplify generate_kmers.py	Calvin

2013-03-05	Added generate_kmers.py	Calvin
	This little script will output all the possible kmer permutations to stdout. The only input is the kmer length desired. Here are some examples. output 2mers ./generate_kmers.py 2 > 2mers.py output N-mers up to 6 with a bit of bash-foo for i in `seq 6`; do ./generate_kmers.py $i > ${i}mers.txt; done;
2013-02-28	use the load from file instead of calling it directly, should work now.	Calvin

2013-02-27	working on the iterators	Calvin

2013-02-27	Create a dictionary of our trained matrix id's	Calvin

2013-02-27	Write the comments in for the work to be done	Calvin
	Comment first, write code later! Also I fixed an erroneous space that really didn't need to be there. Laid down the import for biopyton. Created the basic iteration through our fasta files and added loading of fasta file titles through biopython
2013-02-27	Update multifasta_to_otu.py for new quikr function	Calvin
	* Removed an old comment from the quikr function * Load the trained matrix from multifasta main. * This will reduce load times as we only need to call it once
2013-02-27	* add helper quikr_load_trained_from_file to load file	Calvin
	* change quikr to load a numpy matrix, not file
2013-02-26	added multifasta.m and documenting it	Calvin

2013-02-26	changes	Calvin

2013-02-26	merge resolution	Calvin

2013-02-26	some upodates	Calvin

2013-02-26	iteratively loop through	mutantturkey

2013-02-25	fixed syntax mistake	Calvin

2013-02-25	initial commit, playing around with threading	mutantturkey

2013-02-22	fix example in header, fix weird character in args, remove incorrect dtype ↵	Calvin
	from previous commit
2013-02-22	changed argument help messages	Calvin

2013-02-22	quikr now works! the numbers are almost identical to matlab (is 6th digit ↵	mutantturkey
	precision an issue?). use a vflip and don't transpose and it works
2013-02-22	update readme with link	mutantturkey

2013-02-22	require more arguments since people will always be required to have certain ↵	mutantturkey
	ones, simplfies our checks, also remove all out print shapes
2013-02-22	rotate matricies	Calvin

2013-02-22	Fixed quikr to actually work, insert headers of 1's and 0's into our arrays, ↵	Calvin
	fix a spelling error, writhout our solution vector in a csv formatted as floats