Re: [Transdecoder-users] Using BLASTN evidence for CDS predictions

[Martin M. <mmo...@gm...>]

Hi Alexie,
  hmm, I thought it is calling BLASTP already. I don't think it is worth to come up with yet another
package but I understand your reasonings.
Best,
Martin

Ale...@cs... wrote:
> Hi Martin
> 
> 
>> Yes, I don't mind it would resurrected mistakenly truly dead genes (pseudogenes).
> I think that might be a bit out of the scope of the TransDecoder program as a lot of people would mind. Creating a second software that uses the TransDecoder output and the transcriptome assembly sounds like a better way to do it.
> 
> a
> 
> ________________________________________
> From: Martin MOKREJŠ [mmo...@gm...]
> Sent: Tuesday, 24 June 2014 11:11 PM
> To: tra...@li...
> Subject: Re: [Transdecoder-users] Using BLASTN evidence for CDS predictions
> 
> Martin MOKREJŠ wrote:
>> Hi,
>>   I wonder whether TransDecoder could also run blastn against a somewhat related genome/transcriptome
>> and considered the matching regions. They will be matching exons and the UTRs won't be conserved.
>>
>>   Further, could TransDecoder ignore a STOP codon and merge two ORFs if the macthes are in adjacent
>> frames on the same target strand?
> 
> Aha, forgot to explain a bit more: the alignment to DNA will reveal gaps in either strand so quite helpful
> hint for an artificial insertion/deletion. It should be easy for TransDecoder to realize that at that position
> is likely an error. Yes, I don't mind it would resurrected mistakenly truly dead genes (pseudogenes).
> 
> Martin