counting total bases in a fasta file

[morning latte]

Hello,

I have a fasta file like below.

>NODE_1476_length_303_cov_5.280528
TTAAGTGGGATTTCGTTTAGTGAGGTAGGTACTTTTACTTGGATTTCCATAATTGTATAAG
TCTTTTAGTCGTTTTTGTATTCCTTAGCCAATACATAAGAGTAGGCTTGAGCTAACATTTGA......
>NODE_2306_length_339_cov_2.926254
.
.
.

I want to get total base-pair count from this file. I tried fastx toolkit but it did not work as my fasta format is not what fastx wants (guess fastx wants one line of sequences while mine has two lines of sequences). Could somebody suggest any tool or command for it? Thanks.

[martinghunt]

Here's a quick unix hack:
grep -v ">" file.fasta | wc | awk '{print $3-$1}'

[morning latte]

Thanks, it works very well.

[na_sd]

The number of contigs

I am given a fasta file like this file that you can download it. This file is de novo assembled genome.

I have two very simple questions:

1- How many contigs are present in this file? (my guess: it is equal to the number of rows in the file?!)

2- what is the total genomic length of the assembly?
(my guess: it is 'len=245876' which has been written in the first line)

I am new in DNA sequence analysis stuff, thanks for your kind help.

[GenoMax]

Originally Posted by na_sd

The number of contigs

I am given a fasta file like this file that you can download it. This file is de novo assembled genome.

I have two very simple questions:

1- How many contigs are present in this file? (my guess: it is equal to the number of rows in the file?!)

@Brian and I had answered some of these questions in your Biostars post but looks like you are not convinced. The answers are still going to be the same.

If you do

grep -c "^>" your_file

that will give you a number for the contigs present in your file.

2- what is the total genomic length of the assembly?
(my guess: it is 'len=245876' which has been written in the first line)

I am new in DNA sequence analysis stuff, thanks for your kind help.

That may just be the length of first contig. Does each of your contigs have such a number? We don't know if the contigs you have overlap so it may not be possible to answer this question directly.

PS: I am not going to download the file from dropbox to test.

[na_sd]

Originally Posted by GenoMax

@Brian and I had answered some of these questions in your Biostars post but looks like you are not convinced. The answers are still going to be the same.

If you do

grep -c "^>" your_file

that will give you a number for the contigs present in your file.


That may just be the length of first contig. Does each of your contigs have such a number? We don't know if the contigs you have overlap so it may not be possible to answer this question directly.

PS: I am not going to download the file from dropbox to test.

Thanks.
I am not convinced because my file is not like general fasta file. It has ONLY one line with '>'. The person has asked me about the number of contigs in this file and I am sure that it is greater than 1.
I wish you had downloaded the file and then answered my question.

[GenoMax]

Originally Posted by na_sd

Thanks.
I am not convinced because my file is not like general fasta file. It has ONLY one line with '>'. The person has asked me about the number of contigs in this file and I am sure that it is greater than 1.
I wish you had downloaded the file and then answered my question.

If there is only one line starting with ">" in your file then

a> Your file is either incomplete or
b> the sequence got assembled as a single contig

note: You appear to have removed the file from dropbox.