FastQC analysis

[pari_89]

Hi everyone, I am new to NGS analysis and trying my best to analyse a file of staphylococcus aureus. I would like you to give some suggestions about the quality of data produced by Fastqc concering my images if possible.

Thank you very much .

I have attached the images in a zip file.

[CHObot]

I suggest you post the images inline, as no one in their right mind will click on that attachment.

[pari_89]

hi, they are my analysis and are not found on the web, so I cannot insert the image directly.

Sorry about that but I cannot do otherwise

[Cofactor_DaveMessina]

Hi,

The two plots you include look relatively normal, but the most important one is the per base quality graph.

If you post that one, we could give you a better assessment.


Best,
Dave

[GenoMax]

hi, they are my analysis and are not found on the web, so I cannot insert the image directly.

Sorry about that but I cannot do otherwise

You can use the "go advanced" edit mode when you post. There you will find a "paper clip" like icon that will allow you to upload files from your local computer and attach them to your post inline.

Now a days clicking on a "zip" file on a public forum does seem a bit risky

[pari_89]

Hi,

The two plots you include look relatively normal, but the most important one is the per base quality graph.

If you post that one, we could give you a better assessment.

Here is the per base quality graph as requested

Thank you.

[GenoMax]

Are these Ion torrent reads? Till about 150 bp the median scores are above 25 (e.g. 1 error every ~500 bp) so in that sense this not a great run but it is probably not very bad either.

If you are trying to align to a reference genome you may still be able to use the bases past 150 cycles. If your intended application is something else then you may have to consider trimming the 3' ends.

[pari_89]

Are these Ion torrent reads? Till about 150 bp the median scores are above 25 (e.g. 1 error every ~500 bp) so in that sense this not a great run but it is probably not very bad either.

If you are trying to align to a reference genome you may still be able to use the bases past 150 cycles. If your intended application is something else then you may have to consider trimming the 3' ends.

Yes, these are ion torrent data and I need to use them for assembly. My task is to choose the best assembling tool. So do you recommend me trimming the the low quality reads?? like for e.g Quality score<20.

Thank you.

[GenoMax]

Why not try to align to the available S. aureus genomes to get an idea of what the error rate may be in your reads. If you start finding that the ends of the reads are consistently not aligning then you can start trimming/removing entire reads if they have bad quality overall.

Since this is a bacterial genome you have the luxury of trying assembly multiple ways (trim or not).

This appears to be a good tutorial and should be useful for what you are trying to do.

[pari_89]

Why not try to align to the available S. aureus genomes to get an idea of what the error rate may be in your reads. If you start finding that the ends of the reads are consistently not aligning then you can start trimming/removing entire reads if they have bad quality overall.

Since this is a bacterial genome you have the luxury of trying assembly multiple ways (trim or not).

This appears to be a good tutorial and should be useful for what you are trying to do.

Thank you. I will try to do that but do you think that mapping tools can help as well. Is it better to align reads to a reference or contigs to a reference genome?

Thank you.

[sklages]

You might want to have a look at MIRA for both denovo assembly and mapping of your data. Very well documented, very helpful mailing list and author.

http://sourceforge.net/apps/mediawiki/mira-assembler/

You don't need any extra trimming tool for your data.