[maker-devel] Long introns from Augustus

[Shane Brubaker]

Ok thank you.

-----Original Message-----
From: Carson Holt [mailto:carsonhh at gmail.com] 
Sent: Monday, March 10, 2014 9:14 PM
To: Shane Brubaker
Cc: maker-devel at yandell-lab.org
Subject: Re: [maker-devel] Long introns from Augustus

Maybe.  The max intron length will affect evidence alignments and clustering, which will be used as hints to Augustus. You can give it a try. If you lack transcriptome data, just make sure you provide it with a couple of related proteomes.

--Carson

Sent from my iPhone

> On Mar 6, 2014, at 5:48 PM, Shane Brubaker <sbrubaker at solazyme.com> wrote:
> 
> Actually these are calls directly from Augustus (without using Maker).  They are not purely ab initio in that they are using hints from RNA-Seq data.
> 
> I had noticed that Maker does have some information about max intron length - does that mean it could be taken care of by Maker?  I don't have very good "EST" (transcriptome) assemblies because it is a very difficult organism to sequence.
> 
> 
> -----Original Message-----
> From: Carson Holt [mailto:carsonhh at gmail.com]
> Sent: Thursday, March 06, 2014 3:47 PM
> To: Shane Brubaker; maker-devel at yandell-lab.org
> Subject: Re: [maker-devel] Long introns from Augustus
> 
> Are these the ab intio calls that are merged or final MAKER models.
> 
> —Carson
> 
> 
>> On 3/6/14, 4:41 PM, "Shane Brubaker" <sbrubaker at solazyme.com> wrote:
>> 
>> Hi, we have a very compact genome and we are getting a lot of fused 
>> gene models from running Augustus.  I am wondering if anyone has any 
>> advice about how to prevent introns above a certain cutoff from being created?
>> 
>> I tried a couple of things, some settings in a probabilities file and 
>> also changing a long list of probabilities to another file that 
>> someone had suggested on a forum.  So far I don't really see any changes though.
>> 
>> Any advice would be greatly appreciated.
>> 
>> Thanks,
>> Shane
>> 
>> _______________________________________________
>> maker-devel mailing list
>> maker-devel at box290.bluehost.com
>> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.o
>> rg
> 
>