Errors using evgmran2tsa.pl

Feng Tao

unread,

Jan 28, 2020, 5:48:02 PM1/28/20

to EvidentialGene

Dear Don (and other readers),

I am currently constructing Phragmites transcriptome. As the data constraints, I would like to construct a good transcriptome as I can. I followed your philosophy on this.

So far, I have used four assemblers as Trnitiy, transabyss, shannon and SOAPdenova. I combined them and change the format by trformat.pl script.

Then I followed the instruction to use the tr2aacds.pl to reduce the the combined assembly, and the code as:

perl evigene/scripts/prot/tr2aacds.pl -mrnaseq Phragmites_RNA/all_assemlies/Jan_20_no_cdhit/all_reformat.tr -NCPU=8 -MAXMEM 200000 1>Phragmites_RNA/all_assemlies/Jan_20_no_cdhit/tr2aacds.log 2>Phragmites_RNA/all_assemlies/Jan_20_no_cdhit/tr2aacds.err

I obtained folders as all_reformatnrcd1_blsplit, all_reformat_split, dropset and okset. I also had files as all_reformatnrcd1_db.perf, all_reformat.trclass, all_reformat.trclass.sum.txt.

After this, I am trying to use the evgmrna2tsa2 script as:

perl evigene17dec14/scripts/evgmrna2tsa.pl -onlypubset -idprefix PaustralisEVm -class Phragmites_RNA/all_assemlies/Jan_20_cdhit/all_reformat.trclass

However, some errors occur showing as:

#m2t: EvidentialGene mrna2tsa.pl VERSION 2018.06.18

#m2t: FATAL Missing -mrna .

So my question is which part I did wrong and can you point out and help me to solve it?

As I am still the newbie in this area, although I checked documents listed in this discussion group and some in evidentialGene website. My final purpose is to contruct the transcriptome and compare gene experssion in different tissues and annotate it. Do you have any suggestion what I should keep in mind in the following analyses?

Thanks for your help! Look forwarda to your reply.

Feng

Don Gilbert

unread,

Jan 28, 2020, 9:12:39 PM1/28/20

to Feng Tao, EvidentialGene

Feng,

There may have been an error in results from tr2aacds, or this may be a mix-up for evgmrna2tsa in finding files left by the former.

Your log files may report if tr2aacds had errors finishing.

You can try changing to the directory with the file all_reformat.trclass, which should have a subdirectory okayset/

Check that okayset/ contains large sequence files, likely named all_reformat.okay.aa, .cds, and .tr and all_reformat.okalt.aa, .cds and .tr

Then retry the program evgmrna2tsa.pl .. as per these steps:

cd Phragmites_RNA/all_assemlies/Jan_20_no_cdhit/
ls -l all_reformat.trclass
ls -l okayset/
# do files have data? okayset/all_reformat.okay.aa okayset/all_reformat.okalt.aa okayset/all_reformat.okay.tr ..
# assuming this is path to your evigene folder: $HOME/evigene17dec14/scripts/
# run this command, the -debug -log options add information to log file
perl $HOME/evigene17dec14/scripts/evgmrna2tsa.pl -debug -log -onlypubset -idprefix PaustralisEVm -class all_reformat.trclass

If it works now, there will be a new publicset/ folder containing reformatted sequences and tables of gene locus IDs and information.

- Don Gilbert

--
You received this message because you are subscribed to the Google Groups "EvidentialGene" group.
To unsubscribe from this group and stop receiving emails from it, send an email to evidentialgen...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/evidentialgene/bfe59f24-a6a9-4f9d-9704-97acfb0c04b5%40googlegroups.com.

--

don gilbert - www.bio.net - bioinformatics - indiana.u.

Feng Tao

unread,

Jan 30, 2020, 11:34:04 AM1/30/20

to EvidentialGene

Thanks for kind reply, Don.

My log file regarding the tr2aacds is like that:

# Class Table for Phragmites_RNA/all_assemlies/cdhit/all_reformat.trclass

class %okay %drop okay drop

althi 4.6 6.2 161192 216335

althi1 4 10.1 139198 352626

althia2 0 0.38 0 13367

altmfrag 0.6 0.7 21745 27818

altmfraga2 0.07 0.06 2686 2175

altmid 0.7 0.9 24519 33336

altmida2 0.07 0.04 2594 1620

main 2.8 4.5 98697 156976

maina2 0.17 0.12 6091 4260

noclass 2.4 18.6 85406 649485

noclassa2 0.01 0.08 535 3109

parthi 0 7.9 0 276427

parthi1 0 1.7 0 62438

parthia2 0 0.31 0 11077

perfdupl 0 15.4 0 537639

perffrag 0 16.8 0 586512

---------------------------------------------

total 15.6 84.3 542663 2935200

=============================================

# AA-quality for okay set of Phragmites_RNA/all_assemlies/cdhit/all_reformat.aa.qual (no okalt): all and longest 1000 summary

okay.top n=1000; average=1252; median=1090; min,max=890,5268; nfull=894; sum=1252583; gaps=31,0

okay.all n=190729; average=113; median=72; min,max=30,5268; nfull=130791; sum=21702940; gaps=4955,0

The error file is:

#t2ac: EvidentialGene tr2aacds.pl VERSION 2018.06.18

#t2ac: CMD: tr2aacds.pl -mrnaseq Phragmites_RNA/all_assemlies/cdhit/all_reformat.tr -NCPU=8 -MAXMEM 2000

I do have the all_reformat.trclass and okayset/. Then I rerun the evgmrna2tsa program, and now output files as marnalt.tab, mrna2tsa.info, pudids, pubids.old, pubids.realt.log.

The err file is produced as:

#m2t: EvidentialGene mrna2tsa.pl VERSION 2018.06.18

#m2t: Warn: make_annotab missing publicset/all_reformat.mrna_pub.aa

#m2t: ERR: openRead publicset/all_reformat.mrna_pub.aa

#m2t: ERR: openRead publicset/all_reformat.mrna_pub.cds

At the end, I should have the *.mrna_pub.fa output, right? Which currently I do not have.

Thanks for your help!

To unsubscribe from this group and stop receiving emails from it, send an email to evident...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/evidentialgene/bfe59f24-a6a9-4f9d-9704-97acfb0c04b5%40googlegroups.com.

Don Gilbert

unread,

Jan 30, 2020, 4:21:13 PM1/30/20

to Feng Tao, EvidentialGene

Feng,

I think the problem you found is that output files from tr2aacds.pl are not moved into subdirectories: okayset/ dropset/ inputset/ and tmpfiles/

This is what the -tidy option of tr2aacds.pl does, and evgmrna2tsa.pl is expecting sequence files in those subdirectories. You can find this layout

at this evigene_tr2aacds_test sample:

http://arthropods.eugenes.org/EvidentialGene/plants/arabidopsis/evigene_tr2aacds_test/

arath_TAIR10_20101214up.cdna.gz : input transcript set

arath_TAIR10_20101214up.tr2aacds.log : from -log option
arath_TAIR10_20101214up.trclass : tr2aacds output class table

dropset/ : redundant, dropped transcripts
inputset/ : aa and cds sequence of input
okayset/ : non-redundant sequences
arath_TAIR10_20101214up.okalt.aa.gz
arath_TAIR10_20101214up.okalt.cdna.gz
arath_TAIR10_20101214up.okalt.cds.gz
arath_TAIR10_20101214up.okay.aa.gz
arath_TAIR10_20101214up.okay.cdna.gz
arath_TAIR10_20101214up.okay.cds.gz
tmpfiles/ : work data files

evgmrna2tsa.pl requires these files: arathup.trclass, and the okayset/arathup.okay* and arathup.okalt* sequence files
(but dont need to be .gz, gzipped).

If you move your tr2aacds output sequences into okayset/ subdirectory, like above, then this command may work right:

$evigene/scripts/evgmrna2tsa.pl -onlypubset -idprefix ArathEGr -class arath_TAIR10_20101214up.trclass

I suggest you move other data files in the directory with all_reformat.trclass into another, temporary directory. Including those results of prior attempts with evgmrna2tsa.pl

- Don GIlbert

To unsubscribe from this group and stop receiving emails from it, send an email to evidentialgen...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/evidentialgene/61c7676b-eb84-4e8b-899f-4caea099aa7d%40googlegroups.com.

Feng Tao

unread,

Feb 19, 2020, 3:43:25 PM2/19/20

to EvidentialGene

Hi, Dan. Thanks for your help, I finally obtained the results from evgmrna2tsa.pl. So the products should be aa, aa_pub.fa, cds, cds_pud.fa, mrna, mrna_pub.fa, ann.txt, right? For the downstream analyses and transcriptome assessment, I should use the mrna_pub.fa.

I also checked the transcript amount of the reconstructed transcriptome, are 594317 transcripts too many for the plant?

To view this discussion on the web visit https://groups.google.com/d/msgid/evidentialgene/61c7676b-eb84-4e8b-899f-4caea099aa7d%40googlegroups.com.

Reply all

Reply to author

Forward