ekopylova/deblurring

deblurring

(still in alpha!)

To deblur a directory (with .fasta files), run: clean_indel.sh dirname readlen maxreaderror meanreaderror errorprofile indelprob indelmax

recommended values: maxreaderror - 0.02 meanreaderror - 0.005 errorprofile - 1,0.06,0.02,0.02,0.01,0.005,0.005,0.005,0.001,0.001,0.001,0.0005 indelprob - 0.01 indelmax - 3

note that the directory should contain a single fasta for each sample - so it is the result of split_fasta_on_sample_ids.py

Another option is to run in parallel on a directory using: CleanIndelDirParallel.py dirname -l readlen -e 0.02 -n 50 -m 0.005 -d 1,0.06,0.02,0.02,0.01,0.005,0.005,0.005,0.001,0.001,0.001,0.0005 the -n gives the number of parallel processes to use

note that this curretnly works with qsub!

The main deblurring python script is: CleanSeqsIndel.py

In order to join the deblurred sample .ref.fa files to a single biom table, use: CreateTable.py -d dirname -l -o outputbiomtable it creates a text biom table (outputbiomtable.txt) and the corresponding fasta file (outputbiomtable.seq.fa)