Denovo assembled genome quality

[sanjaysingh765]

I recently assembled a genome using my PE 100 reads and got these datas
N75 336
N50 451
N25 650
Minimum 101
Maximum 6,123
Average 428
Count 171,034
Total 73,189,397





Can anyone tell me that whether they are OK or not?

[westerman]

You could not get your genome published in that form (it is too broken up) so in that aspect the answer is "no". On the other hand you might be able to get some useful information from what you have assembled. You could also deposit the reads and the assembly into SRA or another archive.

[sanjaysingh765]

You could not get your genome published in that form (it is too broken up) so in that aspect the answer is "no". On the other hand you might be able to get some useful information from what you have assembled. You could also deposit the reads and the assembly into SRA or another archive.

Thanks for reply. Actually I want to use this assembled genome for differential expression analysis. Is it OK for that?

[Brian Bushnell]

It's not ideal for that because the fragments are generally shorter than what you would expect for full-length genes. Before using that assembly, I would do whatever possible to improve it, either by better using the existing data (different preprocessing, different assembler, etc) or generating more data.

[westerman]

Usually a person will do rna-seq for differential expression analysis (DEA) and not bother with a whole genome assembly. Of course there are other ways to do DEA. I am not sure what your assembly will get you.

BTW, just as double check, in your first post you were talking about a whole genome project, correct? I ask because will your numbers are poor for a whole genome project they are not bad for a rna-Seq project ... if that is indeed what your biological sample was.

[sanjaysingh765]

Usually a person will do rna-seq for differential expression analysis (DEA) and not bother with a whole genome assembly. Of course there are other ways to do DEA. I am not sure what your assembly will get you.

BTW, just as double check, in your first post you were talking about a whole genome project, correct? I ask because will your numbers are poor for a whole genome project they are not bad for a rna-Seq project ... if that is indeed what your biological sample was.

Actually I assembled this using RNAseq data using CLC genomics workbench. I am a newbie in this field. Am I doing something wrong?

[westerman]

Ah. rna-seq data. That explains it. I do not use CLC myself and unsure if they have a tools for creating denovo transcripts. In my experience there is a big difference in how assembly of a genome and assembly of a transcriptome are approached however in a quick reading of the CLC literature it does not seem like they make such a differentiation.