About Annotation Of Microbial Genome

Question

[wHo]

Hello, everyone. I have a new sequence of microbial genome. However, I am a new hand in annotation. I want to know the process in annotation of microbial genome. I have read some papers about annotation. However, I think it isn't systematic. I want to know the every step of annotation of a microbial genome and the softwares which are used in every step. Thank you very much. I think this question is a question to every new hand. Thx :-D.

[Neilfws]

Guess what. There is no one correct answer :)

[wHo]

Maybe you are right, but I hope you are wrong. :）

[Neilfws]

Trust me, I'm right. There are a number of annotation pipelines but no one, accepted, "standard" solution. And this is true for most problems in bioinformatics. You have to figure out how to make this call yourself.

Answer 1

[Pappu]

Did you look at some literature reporting a new microbial genome? You will find how they annotated the sequence. There are several techniques out there. Glimmer is good for identifying coding regions. Look here: http://openwetware.org/wiki/Wikiomics:Genome_annotation

[wHo]

Ok, I read some paper about annotation, such as Mage, MGAP. They are different. I wonder if there has a standard system which I can use. In that system, I have efficient softwares and analyse data step by step. If you had done this work, you perhaps can give me more suggestion. Thank you for your answer. ^_^

Answer 2

[Konrad]

If there is a close related strain you could use RATT to map the annotations from it to the new one as a starting point.

[wHo]

Ok I will try it. Thank you for your suggestion. ^_^

Answer 3

[Rahul Sharma]

HI,

For annotating the fungal genome, I have used 5 softwares: Exonerate (http://www.ebi.ac.uk/~guy/exonerate/), Augustus (http://bioinf.uni-greifswald.de/augustus/), GlimmerHMM (http://cbcb.umd.edu/software/glimmerhmm/), GeneMark-HMM-ES (http://exon.gatech.edu/) and Evigan (http://www.seas.upenn.edu/~strctlrn/evigan/evigan.html). Please try to install these packages locally on your machine. I generated the hintfiles for Augustus using Exonerate outputs, by mapping the protein sequences from closely related genomes on to the masked assembled genome. Then other set of annotations were generated from GlimmerHMM and Genemark-Hmm-ES. Finally I generated the consensus gene model form these three gene structures using Evigan. In the later stages you can check the intergenic regions using tBLASTn or Exonerate, by mapping again the protein sequences of related genomes to intergenic regions. Hope this would be helpful.

Best wishes, Rahul

[wHo]

Thank you very much. I think bacterial genome is similar to your fungal genome.