[Gmod-help] Re: [maker-devel] [Gmod-gbrowse] Best practices for generating alignments for GBrowse...

[Carson Holt]

>>My basic outline for mapping related genes is to use BLAST for
>>a coarse location, then exonerate for a refined mapping.
>>Using just exonerate w/o pre-located genes is possible but
>>you end up with a much large compute task and generally
>>more gene mappings than you know what to do with.

MAKER runs both blast and exonerate internally so it takes care of this for you.


>>The short answer is
>>if the assemblies don't change much then tools will
>>move annotations from one to the next.  If they do
>>change, you are generally better off re-running your
>>gene-finding, gene-mapping software.

MAKER can do this too. MAKER can run gene-finders like snap, genemark, augustus, fgenesh for you.  In addition, MAKER also produces evidence informed "hints" based on the EST and protein alignments as well as mRNAs from closely related species.  MAKER then talks to the gene predictors using the hints and causes them to produce even better gene models.  MAKER will also revise these gene models to include five and three prime UTRs when there is support from EST evidence.

MAKER will also run other programs for you like RepeatMasker if you choose.  There are also tools for running InterProScan (a protein domain finder), as well as tools for automatically assigning GO functional terms and putative gene functions to each new gene prediction.

The newest revision of MAKER also comes with a handy maker2chado script that will auto-load all MAKER produced GFF3 files into a new or pre-existing chado database for you.

Carson



On 8/14/09 3:31 PM, "Don Gilbert" <gilbertd at cricket.bio.indiana.edu> wrote:




Bob,

Here is my two-bits advice on your general questions of gene to genome
mapping tools.

>> Specifically, we'd like to load the following:
>> ESTs positioned against genomic scaffolds (same species)
>> BLAST? or a better program?

GMAP is the one I use and recommend for EST alignments,
but BLAT works, exonerate, BLAST and others also work.

>> Full-length mRNAs/clones (same species)
>> BLAST? or something better?

add exonerate here (maybe the older GeneWise). exonerate will give
you more complete gene/mRNA mappings than BLAST.

>> mRNAs from closely-related organisms for cross-species comparisons on gene
>> structure
add exonerate here

>> ?
>> proteins from closely-related organisms for cross-species comparisons on
>> gene structure
exonerate

My basic outline for mapping related genes is to use BLAST for
a coarse location, then exonerate for a refined mapping.
Using just exonerate w/o pre-located genes is possible but
you end up with a much large compute task and generally
more gene mappings than you know what to do with.

For your question of remapping to new assemblies, if the
MAKER or Flybase answers are not what you want, look at
UCSC genomes, and Jim Kent's software.  They do a lot
of that, mapping b/n assemblies.  The short answer is
if the assemblies don't change much then tools will
move annotations from one to the next.  If they do
change, you are generally better off re-running your
gene-finding, gene-mapping software.

-- Don

-- d.gilbert--bioinformatics--indiana-u--bloomington-in-47405
-- gilbertd at indiana.edu--http://marmot.bio.indiana.edu/

_______________________________________________
maker-devel mailing list
maker-devel at yandell-lab.org
http://yandell-lab.org/mailman/listinfo/maker-devel_yandell-lab.org


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://brie4.cshl.edu/pipermail/gmod-help/attachments/20090814/a2ee05f1/attachment.html>