gene annotation using BAM files

Question

[djerroudsamia]

Hello everyone,

I am really new in this forum, so I hope you can help me in resolving my problem. I am actually working on a project, where I have to annotate some sequences.these sequences are a result of rna_seq , and I have them in bam format. Do you know some easy bioinformatics tools that can help me to annnotate that kind of sequences in order to get GFF files ?? So the input should be in bam files and the output in GFF. Just for information, i need a tool that can be executed in linux, and that can be looped for so many sequences for different species.

Thank you for your help

[lieven.sterck]

Will need a bit more info to be able to accurately help you here. What is in the BAM file? what kind of species are we talking here? ...

at first sight it seems you are missing a few steps in your 'procedure'

[djerroudsamia]

yes , I admit that, it's the first time I am trying to annotate, so I am completely lost, I got the bam files from the sequencing procedure.do you know all the procedure that I have to do to annotate that kind of files ??

[lieven.sterck]

there is a fair chance I do indeed, but I keep repeating we need much more info.

What exactly do you understand by "annotate"? : find genes, assign functional regions to your BAM file (eg is read aligned in exon , intron, UTR, ... ?), ....

Do you have a genomic sequence to work with?

[djerroudsamia]

My goal is identifying the locations of genes and all of the coding regions in a genome and determining what those genes do. yes i have a genomic sequence

Answer 1

[lieven.sterck]

OK, so you want to annotate (predict genes) on a genomic sequence making use of RNAseq info.

The question is easily asked but I'm afraid you might be underestimating how a complex matter this can be. But is absolute a good and correct approach. Problem is that this often requires quite a bit of work and knowledge. There are (or you could) run most tools as a black box but then your results will be of lesser quality.

You might consider tools as: Maker, Braker, Augustus, ... (most of these use RNAseq as an evidence information but also require other inputs ). There will be others that might work with RNAseq only but none come to mind right now.

[djerroudsamia]

thank you very much your suggestion helped me a lot!!!

[DNAngel]

Hi! Just wondering, does Maker, Braker and Augustus pipeline work for non RNAseq data as well? What about single-end exome sequencing data for non-bacterial and non-mammalian species?

[lieven.sterck]

Most of them will also work without RNAseq data, though keep in mind the accuracy of the results will be severely tempered by not having transcript info