Dear all,
Evidential Gene models for pea aphid are available now at
http://arthropods.eugenes.org/EvidentialGene/pea_aphid2/genes/
Evidential gene sets are built/in-progress for 3 other
arthropods and one plant. If you have an arthropod genome
with lots of new expression/other gene evidence that wants
good gene models, please contact me. I will miss this
week's Arthropod Genomics Symposium at Kansas City, to my
regret. If you run into anyone there with the problem of too
much gene data that is not being turned into good models,
have them contact me.
The software for Evidential Gene is progressing, and is
partly available now, with more work this summer to document
and simplify its use. It can be used by other genome
informaticians.
-- Don Gilbert
Evidential genes for 2 arthropods in 2011
Daphia magna, 2011mar Pea aphid, 2011 june
Evid. Nevd Dmagb8 Evid. Nevd Aphid2
------ ------ ------ ------ ------ ------
EST 26mb 0.86 EST 36mb 0.79
Pro 27mb 0.79 Pro 27mb 0.76
RNA 32mb 0.70 RNA 55mb 0.49
Intron 89k 0.95 Intron 127k 0.70
Coding span 30 mb Coding span 42 mb
Exon span 58 mb Exon span 73 mb
Genome size 131 mb* Genome size 541 mb
Gene count 31694 Gene count 36586
>95% evidence 66% 37%
>66% evidence 81% 66%
-------------------------------------------------
Nevd for EST, Pro, RNA is total span of non-overlapping
reads or alignments, but count for valid unique introns from
spliced reads. * This Daphnia magna genome assembly is
incomplete, and genes for now are limited to members of
the daphnia consortium.
For pea aphid,
36,500 genes are located in aphid2_evigene8f gene set:
14,000 are fully supported by evidence (expression/orthology),
24,000 have above 66% evidence support,
the remainder have evidence but at lower levels.
23,200 have paralogs to pea aphid genes, above 33% identity
13,500 have orthologs to other species, above 33% identity
5000 alternate transcripts add to these primary transcripts.
4400 are non-coding or poorly coding genes.
270 have long genes with valid long introns, span > 100 Kb (acyp2eg0001707t1 nears 1 Mb;
honeybee has 240 long genes, fruitfly has 50).
There are also now NCBI Gnomon/RefSeq for pea aphid assembly 2. Using
my criteria the Evidential gene set is better, if not yet best at
all loci.
Gene Evidence Summary for pea_aphid2, 2011 June
Evid. Nevd Statistic acypi1 evigene ncbi2
------ ------ ------------- ----- ------ -----
EST 36Mb BaseOverlap 0.49 0.79 0.69
Protein 27Mb BaseOverlap 0.47 0.76 0.46
RNA 55Mb BaseOverlap 0.27 0.49 0.43
Intron 127000 SplicesHit 0.52 0.70 0.68
Ortholog SameLocus_N 10127 11114 10157
Ortholog SameLocus_Nbest 537 1370 796
Ortholog SameLocus_Bits 387 402 415
------------------------------------------------------
Predictors are
evigene 2011-June, ncbi2 refseq, 2011-May
acypi1 acyr1-ACYPImRNA, from 2009
EST, Protein, RNA and Intron are as above.
Ortholog SameLocus is protein homology of 3 prediction sets
to Uniprot-arthropods from loci with all 3 predictors:
SameLocus_N is count of models with Uniprot hit (e<1e-5),
SameLocus_Bits is average bitscore,
SameLocus_Nbest is count where predictor has best bitscore by >5%
--------------------------------------------------------------
-- d.gilbert--bioinformatics--indiana-u--bloomington-in-47405
-- gilbertd from indiana.edu--http://marmot.bio.indiana.edu/
