GRAMMy Manual

Commandline Tools

1. grammy_gdt -- to include reference sequences needed by grammy
2. grammy_rdt -- to generate read data needed by grammy
3. grammy_pre -- to parse the alignment results and get the probability matrix needed by grammy
4. grammy_em -- to calculate the estimation(not normalized), the extra bootstrap estimations and the final log likelihood
5. grammy_post -- to normalize the result and get the genome length and relative abundance estimates

grammy_gdt Input

1. o_prefix --output prefix, o_prefix.gdt will be the output filename
2. taxids --taxids to include, taxids: t1,t2,...,tm, each tid is a INTEGER and must be found in grefs/gid_tid.dmp

grammy_gdt Options

usage: grammy_gdt [-h] [-d DMP] [-r REF] [-p PER] o_prefix taxids

optional arguments:
  -h, --help         show this help message and exit
  -d DMP, --dmp DMP  gid to tid dump file, default=grefs/gid_tid.dmp
  -r REF, --ref REF  reference genome dir, default=grefs
  -p PER, --per PER  number of genomes per file, default=20

grammy_gdt Output

1. o_prefix.gdt --genome data file needed by grammy
2. o_prefix.fna.1,...o_prefix.fna.n --including reference sequences needed by grammy

grammy_rdt Input

1. i_prefix --itput dir prefix, a dir where reads files reside
2. o_prefix --output dir prefix, the output will be o_prefix/xxx.rdt, use '.' for current dir

grammy_rdt Options

usage: grammy_rdt [-h] [-s SUF] [-t TEC] [-c CHG] i_prefix o_prefix

optional arguments:
  -h, --help         show this help message and exit
  -s SUF, --suf SUF  read files suffix, default=fa.gz
  -t TEC, --tec TEC  sequencing tech, default=sanger
  -c CHG, --chg CHG  name change set 'o1:n1,o2:n2', default=

grammy_rdt Output

1. o_prefix.rdt --the read data file need by grammy
2. o_prefix.fasta.gz --the zipped reads fasta file

grammy_pre Input

1. read_dat --will use read_dat.rdt as read data file
2. gen_dat --will use gen_dat.gdt as genome data file

grammy_pre Options

usage: grammy_pre [-h] [-m {tbl,bam}] [-p PAR1] [-q PAR2] read_dat gen_dat

optional arguments:
  -h, --help            show this help message and exit
  -m {tbl,bam}, --mtd {tbl,bam}
                        method for read assignment, tbl -- tabular blast format
  -p PAR1, --par1 PAR1  first parameter for read assignment method, tbl filename or k
  -q PAR2, --par2 PAR2  second parameter for read assignment method, al,id,ev or bg

grammy_pre Output

1. read_dat.mtx --the probability matrix file needed by grammy

grammy_em Input

1. read_prob_file --input read probability matrix file from grammy-pre

grammy_gdt Options

usage: grammy_em [-h] [-b BTP] [-t TOL] [-c {U,L}] [-n MIT] [-i {M,R}] read_prob_file

optional arguments:
  -h, --help            show this help message and exit
  -b BTP, --btp BTP     bootstrap number, default=10
  -t TOL, --tol TOL     tolerance for stopping, default=10e-6
  -c {U,L}, --mtd {U,L}
                        convergenece method, (U)niform, (L)ikelihood, default=U
  -n MIT, --mit MIT     maximum number of iteration, default=1000
  -i {M,R}, --ini {M,R}
                        initilization method, (M)oment, (R)andom, default=M

grammy_em Output

1. read_prob_file.est --where the estimation is, not normalized
2. read_prob_file.btp --where the extra bootstrap estimations are
3. read_prob_file.lld --where final log likelihood is

grammy_post Input

1. mix_par --untransformed mixing parameter estimates, from grammy-em
2. gen_dat --gen_dat.gdt genome data file, from grammy-gdt
3. btp --bootstrap file, from grammy-em

grammy_post Options

usage: grammy_post [-h] mix_par gen_dat btp

optional arguments:
  -h, --help  show this help message if exit

grammy_post Output

1. read_prob_file.avl --average genome length estimates"
2. read_prob_file.gra --genome relative abundance, first line is taxon id, second line is relative abundance, last line is error bound"

Have fun!