August 2007 - Mg-rast - mailman3.cels.anl.gov

Fwd: mg-rast Metagenomics data
by Robert Olson 31 Aug '07

31 Aug '07

FYI, on what confused the user. Begin forwarded message: > From: Tom Schmidt <tschmidt(a)msu.edu> > Date: August 31, 2007 4:44:28 PM CDT > To: Robert Olson <olson(a)mcs.anl.gov> > Subject: Re: mg-rast Metagenomics data > > Thanks Bob. That was an easy solution that appears to have worked > - the jobs are now running. > > The reason I had used the .fasta suffix was that in the Upload > page, I interpreted the instructions "If your data set is larger, > use the compressed format ... The fasta file names should end > either *.fna, *.fa, or *.fasta..." to mean that the compressed > file should also have a .fasta suffix. I thought that I had > renamed two earlier compressed files as .fasta, and those worked. > Anyhow, the problem is fixed. > > Bests, > Tom > > Robert Olson wrote: > >> hm. It appears that there may be a bug in the upload processing >> when you submit a file ending in .fasta that is actually a tar >> file. Try uploading using a file with a .tgz suffix instead. I'll >> check on making the server do a better job of handling this case. >> >> --bob > > > > -- > ******************************************* > Tom Schmidt, Professor > Dept. Microbiology & Molecular Genetics > 6180 Biomedical and Physical Sciences Bldg. > Michigan State University > East Lansing, MI 48824-4320 > > Phone: (517) 355-6463 x1606 > Fax: (517) 353-8957 > Email: tschmidt(a)msu.edu > ******************************************* > >

1 0

mg-rast Error on uploads in mgrast
by Robert Olson 31 Aug '07

31 Aug '07

If one uploads a file foo.fasta that is actually a gzipped tarfile, it looks like the front end doesn't recognize this (the source_archive key does not get set. See job 170 as an example. --bob

1 0

mg-rast how to make a tar compressed file
by David Mead 23 Aug '07

23 Aug '07

Hi, I would like to use your metagenomic analysis software on the web and I can not figure out how to make a tar compressed file. I have a Windows 2000 PC. Any help is appreciated. Thanks, David Mead, Ph.D. Lucigen Corp. 2120 W Greenview Dr Middleton WI 53562 T: 608-831-9011 ext 245 F: 608-831-9012 email: dmead(a)lucigen.com www.lucigen.com

1 0

Re: mg-rast Metagenomics data
by Robert Olson 17 Aug '07

17 Aug '07

Good afternoon - My apologies for the confusion. The site you need to go to is http:// metagenomics.nmpdr.org/ ; I have corrected the software that generates the email message as well. --bob On Aug 17, 2007, at 12:24 PM, Tom Schmidt wrote: > Hello, > > I uploaded some 454 metagenomic data into MG-RAST and received > email notification (copied below) that the analysis was completed. > However after looking around the web site, I still can't find the > results. Any help you can offer would be appreciated. Thanks for > making this analysis possible - I will let you know how things turn > out with this particular analysis of a microbial community from soil > > Tom Schmidt > > "The annotation job that you submitted for MA1W2 has completed. > It is available for browsing at http://www.nmpdr.org/anno-server/ > as job number 156." > > ******************************************* > Tom Schmidt, Professor > Dept. Microbiology & Molecular Genetics > 6180 Biomedical and Physical Sciences Bldg. > Michigan State University > East Lansing, MI 48824-4320 > > Phone: (517) 355-6463 x1606 > Fax: (517) 353-8957 > Email: tschmidt(a)msu.edu > ******************************************* > >

1 0

mg-rast Metagenomics data
by Tom Schmidt 17 Aug '07

17 Aug '07

Hello, I uploaded some 454 metagenomic data into MG-RAST and received email notification (copied below) that the analysis was completed. However after looking around the web site, I still can't find the results. Any help you can offer would be appreciated. Thanks for making this analysis possible - I will let you know how things turn out with this particular analysis of a microbial community from soil Tom Schmidt "The annotation job that you submitted for MA1W2 has completed. It is available for browsing at http://www.nmpdr.org/anno-server/ as job number 156." ******************************************* Tom Schmidt, Professor Dept. Microbiology & Molecular Genetics 6180 Biomedical and Physical Sciences Bldg. Michigan State University East Lansing, MI 48824-4320 Phone: (517) 355-6463 x1606 Fax: (517) 353-8957 Email: tschmidt(a)msu.edu *******************************************

1 0

mg-rast Fwd: (Case 575) Job #393
by Folker Meyer 16 Aug '07

16 Aug '07

Hi, there seems to be two issues: a) user experiencing problems with export. b) error reports not directed to the appropriate address (which is mg- rast(a)mcs.anl.gov) Daniel can you work on this with Bob? Folker Begin forwarded message: > From: "Awdhesh Kalia" <a0kali02(a)louisville.edu> > Date: August 16, 2007 8:47:26 AM CDT > To: "Leslie Klis McNeil" <help(a)nmpdr.org> > Cc: <folker(a)mcs.anl.gov> > Subject: Re: (Case 575) Job #393 > > Hello again, > > Have you been able to localize and resolve the glitch? > Additionally, the two other genomes that I had annotated previously > and had fine gbk export files, are now exporting similar erroneous > formats. If you could let me know when this problem is likely to be > resolved that would be grand. > > Best regards, > > Awdhesh > > >>> "Leslie Klis McNeil" <help(a)nmpdr.org> 08/14/07 9:41 AM >>> > Hello, > > I have attempted to download your genome (job 393), and I get only > the first 1361 features in the genbank and embl files, with no > DNA. All 3482 pegs and the rnas appear in the features table > within the SEED viewer, so there must have been a glitch in output > of the annotation files. > > I have copied this reply to Folker Meyer, who will be able to help > redo the files for export. > > thanks, Leslie McNeil > > > -----Original Message----- > From: "Awdhesh Kalia" <a0kali02(a)louisville.edu> > Date: Mon, 13 Aug 2007 17:53:22 -0400 > To: <help(a)nmpdr.org> > Subject: Job #393 > > >Hi, > > > >I have annotated a Legionella longbeachae genome a few days ago > (Job #393) and I have a problem in downloading the whole annotated > genome in GenBank (or EMBL) format. The basic problem is that the > annotated genome file is missing nucleotide data information > altogether. As a consequence I am unable to do further work using > these files (eg., whole genome alignments). I have annotated two > additional genomes and I have never encountered this problem > before. I will greatly appreciate if you can help me resolve this > at your earliest convenience. > > > >Best regards, > > > >Awdhesh Kalia > > > > > >Awdhesh Kalia, Ph.D > >Assistant Professor > >Department of Biology and of > >Microbiology and Immunology > >University of Louisville > >Louisville, KY 40292 > > > >Phone: 5028520991 > > > > > >

1 0

mg-rast Explanation of numbers
by Tobias Paczian 15 Aug '07

15 Aug '07

Hi all, I have started a public wiki page explaining the numbers in the SeedViewer, RAST Server and MG-RAST. Please take a look at it and correct / expand it at your leisure. Also, this needs to be pointed at from a prominent place. http://www.theseed.org/www.theseed.org/wiki/index.php/ SEED_Viewer_Numbers Cheers Tobi

1 0

mg-rast Fwd: RAST-Server
by Tobias Paczian 09 Aug '07

09 Aug '07

Hi all, this is from a user of the metagenomics server. He uploaded two jobs yesterday: 148 alanzen 4440127.3 RudSoil-blastx30bit 150 alanzen 4440128.3 RudSoil-Archaea They seem fine to me, but maybe we should look into why he had the 'could not write data on disk' error. Could this have anything to do with the server-down yesterday night? Cheers Tobi Begin forwarded message: > From: "Anders Lanzï¿½n" <anders.lanzen(a)gmail.com> > Date: August 9, 2007 10:13:56 AM CDT > To: "Tobias Paczian" <paczian(a)mcs.anl.gov> > Subject: Re: RAST-Server > > Hi, > > So far it seems to work. I had to resubmit a few times though, as > it told me it could not write my data on disk, but just doing a > reload seems to have worked. You might want to look into that thoug. > > Cheers, > Anders >

1 0

mg-rast Fwd: Questions
by Robert Olson 09 Aug '07

09 Aug '07

Begin forwarded message: > From: Torsten Thomas <t.thomas(a)unsw.edu.au> > Date: August 8, 2007 11:36:15 PM CDT > To: Robert Olson <olson(a)mcs.anl.gov> > Subject: Re: Questions > > Hi Bob. > > Thanks for the reply. > > I will check with Rob about details regarding the ORF calling (hope > my questions didn't cause too much trouble). Is the relatively low > blastx cut-off of 0.1 only used to identify an putative ORF or also > to assign the ORF a functional hit to SEED? If the first, what is > the cut-off then for functional assignment. > > My NR blast is quite large, but I can email you a subset of > results. Which of the samples I submitted to MG-RAST would you like > to have? > > Best wishes, > > Torsten > > > Dr. Torsten Thomas > Senior Research Fellow > Centre for Marine Biofouling and Bio-Innovation > School of Biotechnology and Biomolecular Sciences > The University of New South Wales > Sydney, NSW 2052 > Australia > Ph + 61 (0) 2 938 53467 > Mob +61 (0) 410159525 > Fax + 61 (0) 2 938 51779 > > > On 04/08/2007, at 9:04 AM, Robert Olson wrote: > >> Hi -- >> >> Rob Edwards put together the ORF caller based on the blastx hits, >> so he knows the details better than I; I haven't yet studied what >> it's doing in detail. The blastx sims are computed with an e-value >> cutoff of 0.1. It appears to create the longest-possible ORF from >> any overlapping hits in the input data. >> >> I would be curious to see some areas in the data where the MG-RAST >> caller found coding regions and your analysis did not, in order to >> see what exactly is happening (I must admit that your question has >> led to a long discussion on the issue of ORF calling in the MG- >> RAST :-). >> >> The counts add up to more than the number of sequences because >> there are sequences that appear in multiple subsystems and hence >> multiple categories. For instance, the called region fig| >> 4440074.3.peg.1074 in your data was a best hit to the SEED CDS fig| >> 237727.3.peg.1240 which participates in four different subsystems, >> in four categories: >> >> Serine-glyoxylate cycle: One-carbon Metabolism >> TCA Cycle: Central carbohydrate metabolism >> Succinate dehydrogenase: Electron donating reactions >> YgfZ (unclassified, it is a bug that this occurs in the count) >> >> --bob >> >> >>>> I submitted about 11 000 shotgun Sanger reads of a marine, >>>> microbial sample. About 9000 coding sequences were identified, >>>> of which more than 7000 (appr. 84%) were classified as non- >>>> hypotheticals. My first impression was that is constitutes a >>>> relative high proportion of hits when compared to our blastX >>>> analysis against NR. Here we have gotten for the same dataset >>>> only hits for about 45% of the reads with a (very loose) cut-off >>>> of 25% identity over at least 25 amino acid positions. >>>> >>>> In order to understand better what the RAST output means I would >>>> like to know, how you identify your coding regions and what cut- >>>> off you set for your search against your SEED database? >>>> >>>> I also noted the sum of the numbers of hits per category is also >>>> much higher than the number of coding sequence detected. Does >>>> this mean that a coding sequence can be assigned to multiple >>>> SEED categories? >

1 0

Re: problem with mg-rast
by Robert Olson 03 Aug '07

03 Aug '07

it comes up, taking a lot of time and memory in the process. It appears to be reading the contigs into memory; not sure what is going on there. It's also connecting to wikipedia.org and getting / wiki/Hib2k4 ? On Aug 3, 2007, at 1:19 PM, Folker Meyer wrote: > Hi all: > > this page does not load: > > http://metagenomics.theseed.org/index.cgi? > action=ShowOrganism&initial=1&genome=4440111.3&job=128 > > > Folker and Liz >

1 0