Scripts Calculate Tetranuclear Frequency

Question

[Shuixia100]

Hi there,

Does anyone as a scripts for tetranucleotide frequency calculation? I have tried to use the BioStrings in Bioconductor but Im looking for something that could calculate the tetranucleotide frequency from both forward and reverse trands simultaneously and input into ESOM Tools for binning of metagenomic contigs.

plus any suggestion on metagenomic binning is greatly welcomed.Is there a established program to do the binning ? Ive tried TETRA, it is just too slow for NGS sequences.

THX Kylie

Answer 1

[Leonor Palmeira]

In R, with seqinr:

library(seqinr)
seq=read.fasta("sequence.fa")[[1]]
count(seq,freq=TRUE,wordsize=4) # for the tetranucleotide frequency (edited)
count(rev(comp(seq)),freq=TRUE,wordsize=4) # but this is exactly the reverse complement of the previous vector (if you see what I mean)

[Michael]

Hi, I think op is asking for counts for all 4-mer words (tetra-nucleotides), I think you have to use wordsize=4 as parameter for count.

[Leonor Palmeira]

Sorry, I meant to do that, and then I forgot...

[Shuixia100]

Then what would be the scripts, plus do you have some documentation for beginers to seqinr?

[Leonor Palmeira]

Are you unable to click on the link I provided or to read my updated code in my answer?

[Shuixia100]

sorry dont notice the update THX

Answer 2

[Pawel Szczesny]

As for binning, there's a recent paper entitled A comparative evaluation of sequence classification programs . Authors assessed not only sensitivity/precision but also runtime performance.