lenght distribution range in a a fasta file

[antoza]

Hi folks,

I have to calculate the lenght distribution range in a a fasta file, for example how many sequences are less than 100 bp, how many are in their lenght from 101 to 300 bp, from 301 to 500 bases and so on.. Any script or tool for doing this job?

Thanks in advance!

[dpryan]

If you index it with "samtools faidx" the resulting ".fai" file will be a text file containing the length of each of the sequences (among other information). You could then plot the distribution in R with whatever binning strategy you want.

[antoza]

Dear dpryan,

I have retrieved the ".fai" file and I have the lenghts of the genes that I wanted (2 columns file, first column are the gene names, second column the gene lenghts, as in the following):

gene_120397 43056
gene_240653 224380
gene_150423 68254
gene_143456 10090
gene_141140 15291
gene_253613 3088

Could you please indicate me an R code for plotting these distributions as I am not so familiar with plotting in R

Thank you!!!

[rnaeye]

mydata <- read.table("inputfile.txt")
plot(mydata)

[arcolombo698]

FastQC Length Distribution

Hello.

I believe that Fastqc has this information, but not in a fasta file.

if anyone has used FastQC to find the length distribution, I am wondering in what conditions does it consider the distribution to be a fail or a pass.

advice?

[westerman]

Hello.

I believe that Fastqc has this information, but not in a fasta file.

if anyone has used FastQC to find the length distribution, I am wondering in what conditions does it consider the distribution to be a fail or a pass.

advice?

From the documentation:

Warning

This module will raise a warning if all sequences are not the same length.

Failure

This module will raise an error if any of the sequences have zero length.

[kurban910]

hi @antoza,
did you find way to solve your problem? if u did could you share your experience?
i am facing with the same task right now.
thanks.

[arcolombo698]

FastQC does plot this information so one can visually see the distribution of length. this is a quick/easy approach.

if you use samtools, one can plot the lengths using R.

[Brian Bushnell]

The BBMap package has a couple programs for this purpose:

stats.sh in=file.fasta shist=shist.txt
(only works on fasta input)

readlength.sh in=file.fasta out=hist.txt
(works on fasta, fastq, or sam)

The way they display output is a little different, but both are easy to plot.

[kurban910]

yeah , that will work. thanks @Brian Bushnell

[kurban910]

hi @arcolombo698,
could you be little bit more specific?!
how could i do what i want by using FastQC when my input file is fasta?

[kurban910]

hey @arcolombo698,
i sued samtools got the fai file. and when i tried to do the length distribution by R i changed my gene.fa.fai file to gene.txt. then i used this commend as its mentioned by@rnaeye:

mydata <- read.table("gene.txt")
plot(mydata)

and got this error:

> mydata <- read.table("gene.txt")
> plot(mydata)
Error: cannot allocate vector of size 156.2 Gb

i am new at R, so could u explain to me where did it go wrong?