April 2011 - Mg-rast - mailman3.cels.anl.gov

Re: [mg-rast] protein sequence extraction using Workbench
by William Trimble 25 Apr '11

25 Apr '11

There was something wrong with what the pipeline did to this dataset. 900.loadDB.sims.filter.seq contains 113948 lines, of which 105964 start with the fastaid 999995. These lines aren't all unique and correspond to 25947 unique md5s. The sims file does not contain any hits for the feature, 999995_1_71_- or the cluster aa90_296467. The mapping file contains the fastaid only once, as expected: aa90_296467 1792719_1_88_+ 1424717_1_78_-,1759585_1_87_-,999995_1_71_-,1259169_1_75_+,1779975_1_88_+ 92.0%,92.9%,100.0%,100.0%,100.0% So the basic symptom: almost all the fastaid's in the loaddb table are identical and don't seem to correspond to a sequence with any real similarites. W On Mon, Apr 25, 2011 at 10:21 AM, Mark D'Souza <dsouza(a)mcs.anl.gov> wrote: > Hi, > > We are looking into the workbench download problem and will get back to you as soon as this is resolved. > > ----- Original Message ----- >> From: "David Bernick" <dlbernick(a)gmail.com> >> To: mg-rast(a)mcs.anl.gov >> Sent: Monday, April 25, 2011 12:23:55 AM >> Subject: [mg-rast] protein sequence extraction using Workbench >> Greetings - >> >> It seems that the fasta sequences returned using the workbench, >> download fasta tool, is returning sequences that do not blast to the >> correct species. The fasta sequences are also nucleic acid rather than >> protein, and their length appears to be fixed. >> >> What's up? >> >> David Bernick >> UCSC >> >> here is an example: > > -- > Regards, > Mark D'Souza > -- for the MG-RAST team > > mg-rast(a)mcs.anl.gov > http://metagenomics.anl.gov/ > > If you need to respond to this email please address it to the mailing-list mg-rast(a)mcs.anl.gov >

1 0

Re: [mg-rast] Normalization
by Mark D'Souza 25 Apr '11

25 Apr '11

Hi Marian, I will pull together the answers and get back to you soon. Regards, Mark ----- Original Message ----- > From: "Marianyoly Ortiz" <moo(a)email.arizona.edu> > To: mg-rast(a)mcs.anl.gov > Sent: Sunday, April 24, 2011 3:43:56 PM > Subject: [mg-rast] Normalization > Hello, > > Thank you very much for your help with my samples. They are all done! > > I have some more questions about data interpretation for you, hope not > too many. > > 1. Can you explain how to interpret the normalization results? I read > your explanation in the website but it wasn't completely clear for me. > The reason is because when I look at the normalized data I don't > understand what the values mean. For example, when I look at the raw > data it shows that Bacteria is >99% and viruses <1% but when > normalized Bacteria is 0.92 and viruses 0.23? What does this value > mean and why it does not add to one? > > 2. Is there a way I can grab the sequences that have specific hits (i. > e. viral hits, prophage hits) and download them to look at them > further? > > 3. When doing taxonomic classification we get multiple hits per read, > correct? But when functional assigments, do we get the best hit? > > Thanks in advance, > > Marian > > -- > Marianyoly Ortiz-Ortiz > Graduate Associate, Research > Soil, Water & Environmental Science > The University of Arizona > Tucson, AZ 85721 > FCS 321 > 520-621-9759

1 0

Re: [mg-rast] protein sequence extraction using Workbench
by Mark D'Souza 25 Apr '11

25 Apr '11

Hi, We are looking into the workbench download problem and will get back to you as soon as this is resolved. ----- Original Message ----- > From: "David Bernick" <dlbernick(a)gmail.com> > To: mg-rast(a)mcs.anl.gov > Sent: Monday, April 25, 2011 12:23:55 AM > Subject: [mg-rast] protein sequence extraction using Workbench > Greetings - > > It seems that the fasta sequences returned using the workbench, > download fasta tool, is returning sequences that do not blast to the > correct species. The fasta sequences are also nucleic acid rather than > protein, and their length appears to be fixed. > > What's up? > > David Bernick > UCSC > > here is an example: -- Regards, Mark D'Souza -- for the MG-RAST team mg-rast(a)mcs.anl.gov http://metagenomics.anl.gov/ If you need to respond to this email please address it to the mailing-list mg-rast(a)mcs.anl.gov

1 0

Fwd: protein sequence extraction using Workbench
by William Trimble 25 Apr '11

25 Apr '11

MGRID 4461888.3 Job 21485, has 1M sequences varying in length from 64 bp to 21861bp, average length 115. Output from an assembler. Prefix diversity is consistent with shotgun. Workbench export contains 15900 sequences of which only 1203 are unique, and all are length 71, 72, or 73, but 14697 of those are occurrences of the same sequence >4461888.3|999995 This is puzzling. W ---------- Forwarded message ---------- From: David Bernick <dlbernick(a)gmail.com> Date: Mon, Apr 25, 2011 at 12:23 AM Subject: [mg-rast] protein sequence extraction using Workbench To: mg-rast(a)mcs.anl.gov Greetings - It seems that the fasta sequences returned using the workbench, download fasta tool, is returning sequences that do not blast to the correct species. The fasta sequences are also nucleic acid rather than protein, and their length appears to be fixed. What's up? David Bernick UCSC here is an example:

1 0

protein sequence extraction using Workbench
by David Bernick 25 Apr '11

25 Apr '11

Greetings - It seems that the fasta sequences returned using the workbench, download fasta tool, is returning sequences that do not blast to the correct species. The fasta sequences are also nucleic acid rather than protein, and their length appears to be fixed. What's up? David Bernick UCSC here is an example:

1 0

Normalization
by Marianyoly Ortiz 24 Apr '11

24 Apr '11

Hello, Thank you very much for your help with my samples. They are all done! I have some more questions about data interpretation for you, hope not too many. 1. Can you explain how to interpret the normalization results? I read your explanation in the website but it wasn't completely clear for me. The reason is because when I look at the normalized data I don't understand what the values mean. For example, when I look at the raw data it shows that Bacteria is >99% and viruses <1% but when normalized Bacteria is 0.92 and viruses 0.23? What does this value mean and why it does not add to one? 2. Is there a way I can grab the sequences that have specific hits (i. e. viral hits, prophage hits) and download them to look at them further? 3. When doing taxonomic classification we get multiple hits per read, correct? But when functional assigments, do we get the best hit? Thanks in advance, Marian -- Marianyoly Ortiz-Ortiz Graduate Associate, Research Soil, Water & Environmental Science The University of Arizona Tucson, AZ 85721 FCS 321 520-621-9759

1 0

Re: [mg-rast] [mgrastdev] server error
by Folker Meyer 23 Apr '11

23 Apr '11

Dear Rory, thanks for bringing this to our attention. We are aware of the problem and are working on it. Best, Folker On Apr 23, 2011, at 5:20 AM, Rory Welsh wrote: > I was attempting to work on my data just now and received the following error: > > Internal Server Error > > The server encountered an internal error or misconfiguration and was unable to complete your request. > > Please contact the server administrator, mgrastdev(a)mcs.anl.gov and inform them of the time the error occurred, and anything you might have done that may have caused the error. > > More information about this error may be available in the server error log. > > > > Any information on when this issue will be resolve would be greatly appreciated. > > > > Many Thanks, > > Rory > > > -- > Florida International University > Department of Biological Sciences > Office MB-250F > 3000 NE 151 Street > North Miami, FL 33181 > (305) 919-4109 > (305) 919-4030 FAX > _______________________________________________ > mgrastdev mailing list > mgrastdev(a)mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mgrastdev

1 0

Re: [mg-rast] [mgrastdev] MG-RAST down
by Folker Meyer 23 Apr '11

23 Apr '11

Hi Donald, thanks for bringing it to our attention. We are aware of the problem and are working on it. Best, Folker On Apr 23, 2011, at 5:07 AM, dpan(a)huskers.unl.edu wrote: > Hi, MG-RAST seems to be down at around 9:40pm central time. > > > ================================================================== > Donald Pan > PhD student, Biological Sciences - Microbiology & Molecular Biology > Microbiology Graduate Student Association Vice President > University of Nebraska Lincoln > > Weber Microbial Biogeochemistry & Geomicrobiology Lab > Office: 232.1 Manter Hall > Laboratory: 207,234 Manter Hall > Phone: 402-472-0154 > E-mail: dpan(a)huskers.unl.edu > Website: http://lifetheuniverseandgradschool.blogspot.com > ================================================================== > _______________________________________________ > mgrastdev mailing list > mgrastdev(a)mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mgrastdev

1 0

Re: [mg-rast] MG-RAST
by Gary Xie 22 Apr '11

22 Apr '11

Hi, Mark, I am very impressed by all rich features of V3 provided. The video training is very helpful for me to switch everything from V2. I have noticed one of my old samples has been in progress for quite long time. Would you check it out? • Metagenome ID - Name4453523.3 - Saliva SRS015055 • Job13129 Thanks, Gary On Apr 14, 2011, at 12:56 PM, Mark D'Souza wrote: > Gary, > > 12132 has been restarted and 21067 halted. We will make the delete > job available again at some point. > > Regards, > Mark > > ----- Original Message ----- >> From: "Gary Xie" <xie(a)lanl.gov> >> To: "Mark D'Souza" <dsouza(a)mcs.anl.gov> >> Cc: mg-rast(a)mcs.anl.gov >> Sent: Thursday, April 14, 2011 12:44:44 PM >> Subject: Re: MG-RAST >> Thanks, you can just delete the job 21067. I remember the V2 had a >> feature allowing user to delete the job. >> Gary >> >> On Apr 14, 2011, at 11:35 AM, Mark D'Souza wrote: >> >>> Hi Gary, >>> >>> If you would like the metagenome name changed I can take care of >>> that. >>> >>> 12132 was scheduled to be run on Version 3 but never made it >>> through, I will check what happened. >>> >>> Mark >>> >>> >>> ----- Original Message ----- >>>> From: "Gary Xie" <xie(a)lanl.gov> >>>> To: "Mark D'Souza" <dsouza(a)mcs.anl.gov> >>>> Cc: mg-rast(a)mcs.anl.gov >>>> Sent: Wednesday, April 13, 2011 11:53:29 AM >>>> Subject: Re: MG-RAST >>>> HI, Mark, >>>> Thanks for clarifying. >>>> >>>> I just submitted this data for annotation. BTW, Would you delete >>>> the >>>> job Jobs #21067? I made a mistaken on its title. >>>> And how about the Jobs #12132? Is it going to be processed? >>>> Thanks, >>>> Gary >>>> >>>> >>>> >>>> >>>> On Apr 13, 2011, at 8:42 AM, Mark D'Souza wrote: >>>> >>>> >>>> >>>> Hi Gary, >>>> >>>> The upload/job creation is a two step process, your upload of the >>>> 1000091.1.clean.trimmed.fasta file appears to have completed fine. >>>> The >>>> system then does some preprocessing steps before returning with the >>>> job creation step. When you go to the upload page you will see a >>>> 'Create Job' tab which contains information about the sequences in >>>> the >>>> file you submitted and a form where you can select the QC options >>>> and >>>> create the job. >>>> >>>> Regards, >>>> Mark >>>> >>>> ----- Original Message ----- >>>> >>>> >>>> From: "Gary Xie" < xie(a)lanl.gov > >>>> >>>> >>>> To: mg-rast(a)mcs.anl.gov , "Gordon D. Pusch" < gdpusch(a)xnet.com >, >>>> "Mark D'Souza" < dsouza(a)mcs.anl.gov > >>>> >>>> >>>> Sent: Tuesday, April 12, 2011 6:47:31 PM >>>> >>>> >>>> Subject: MG-RAST >>>> >>>> >>>> Hi, Gordon and Mark, >>>> >>>> >>>> I had several metagenome samples processed by MG-RAST v2 few months >>>> >>>> >>>> ago. Right now, one additional sample (one lane illumina 76bp >>>> reads) >>>> >>>> >>>> needs to be annotated for the comparison purpose. >>>> >>>> >>>> >>>> >>>> >>>> Just went to your site and noticed that V2 no longer accept new >>>> >>>> >>>> annotation. The V3 allowed me to upload the sequence file, but it >>>> >>>> >>>> didn't ask any question regarding how it will be annotated. Am i >>>> >>>> >>>> missing anything here? >>>> >>>> >>>> Thanks, >>>> >>>> >>>> Gary

1 0

Re: [mg-rast] different results in version 2 and 3 of MG-RAST
by Mark D'Souza 22 Apr '11

22 Apr '11

Hi, As you mentioned, there are some changes to be expected due to the updated databases and changes in the sequence analysis algorithm. The numbers being displayed for a single database source (as in the 90 for SEED) in v3 are currently misleading due to the way the hits are being counted, selecting M5NR as the database source will give results that are more appropriate to compare with the v2 numbers. We are aware that this behavior is counter-intuitive and will be changing this in the future, thanks for bringing it to our attention. Regards, Mark D'Souza -- for the MG-RAST team mg-rast(a)mcs.anl.gov http://metagenomics.anl.gov/ If you need to respond to this email please address it to the mailing-list mg-rast(a)mcs.anl.gov On Apr 22, 2011, at 3:19 AM, Laura FANCELLO wrote: > Hi, > I'm a phd student and I'm using your new version of MG-RAST on my metagenomes. Actually I realized that the same analysis performed by MG-RAST v2 And by MG-RAST v3 doesn't give the same results. I'm performing phylogenetic annotation on metagenomes 4453567.3, 4453569.3, 4453570.3 against the SEED database with an e-value<1e-05 and I noticed that the number of sequences annotated as Enterobacteriaceae or Streptococcaceae or pseudomonadaceae > is absolutely not the same: for example for metagenome 4453567.3 I find 6292 enterobacteriaceae in version 2 and 90 in version 3. I wonder if it is due to an error or rather to a different way to perform the blast search or to an update/new curation of the SEED database... I know that I have a lot od beta-lactamase belonging to cloning/expression vector sequences in my metagenome and maybe these vectorial sequences are considered as bacterial in version 2 and as artificiail sequences in version 3 but this wouldn't explain the differences in the other cases (Burkholderia, Streptococcus, Staphylococcus)... > Can you help me?? > Thank you very much!!!! > Laura > > > -- > > Laura FANCELLO > Universite de la Mediterranee (Aix-Marseille II) > http://annuaire.univmed.fr/showuser.php?uid=fancello.l

1 0