name-that-microbe

The idea is to make a version of probeseq that works with not only oral bugs but any kind of reference database to quickly identify 16S sequences from a miseq experiment.

V regions based on E. coli annotation

Region	Start	End
V1	69	99
V2	137	242
V3	338	533
V4	576	682
V5	822	879
V6	967	1046
V7	1117	1173
V8	1243	1294
V9	1435	1465

References used

• gg_otus-13_8-release/rep_set/97_otus.fasta
• NCBI's 16S database (to be added)

Run a small test dataset

Generate a reference database

cd ~/repos/name-that-microbe

IN=tests/small.fasta
OUT=tests/small.out
LOG=tests/small.log
AMBIG=tests/small_ambig.fasta

#V1-3 region
#START=69
#END=533

#V3-4 region
START=338
END=682

python utils/generate_ref_set.py -i $IN -o $OUT -s $START -e $END -l $LOG -a $AMBIG

Annotate some fake sequence data

cd ~/repos/name-that-microbe

REF_FILE=tests/small.out
FASTA=tests/small_dataset.fasta
OUTFILE=tests/small_dataset_annotated.txt
LOG=tests/small_dataset.log

python annotate.py -r $REF_FILE -i $FASTA -o $OUTFILE -l $LOG