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

Uploaded files still in progress step: processing upladed file...
by Trine Frisli 26 Aug '11

26 Aug '11

Hi, I tried to upload some files on your web page some days ago, and they are still in the progress step: processing uploaded file. The files are in fasta format and not very big (around 600-700kb). I am not able to create jobs from these files. Two of the files that were loaded up, seemed to be processed right away, and I were able to create jobs, but I do not see the in the in progress under your data summary. Are there something wrong with my files? I have used the same types of files before, and everything went fine then. Hope you can answer me and tanks for reading my mail. Best regards from Trine Frisli Trine Frisli PhD-student Biotechnology Department of Natural Sciences and Technology Faculty of Education and Science Hedmark University College P.O.Box 4010, N-2306 Hamar Norway Phone: +4762517856

1 0

sequence names in MGRAST
by hakon.dahle＠bio.uib.no 26 Aug '11

26 Aug '11

Hi I am a user of MGRAST and I find this application to be a great tool in analyses of my transcriptomic data. But I have one question: In the annotation table I would have liked to see the sequence names of the sequences I uploaded. Is this possible somehow? (I need this for comparing annotations from various resources on a sequence by sequence level). Best H

1 0

Re: [mg-rast] # reads hit
by Mark DSouza 25 Aug '11

25 Aug '11

Hi, Sorting by '# reads hit' seems more suitable as it will give you the genes in the reference organism which are most abundant in the metagenomic dataset. It is based purely on the similarity information and the numbers in each row will include effects due to number of organisms with similarity and gene copy number. While this will give you the more abundant genes, I don't know how well this will correlate with the importance of the function for the survival in an environment. The numbers next to the organism name in the reference organism listing represents the number of hits against that organism, sorted in descending order so that the organism at the top generally displays greater similarity to the metagenomic data than the one at the bottom, especially when the difference in the number of hits is large. You are welcome, glad to be of help, Mark ----- Original Message ----- > From: "Joanne Wyglinski, Ms" <joanne.wyglinski(a)mail.mcgill.ca> > To: "Mark DSouza" <dsouza(a)mcs.anl.gov> > Sent: Thursday, August 25, 2011 11:57:24 AM > Subject: RE: [mg-rast] # reads hit > Dear Mark, > > Thank you for your response, it is greatly appreciated. > > I am looking for which genes in relevant organisms in the reference > set (ie. thermophiles) are the most abundant in my sample so that I > can figure out which genes play the most important role in terms of > how these organisms survive or thrive in the environment that they are > in. Do the # reads hit take into account the number of organisms with > similarity or gene copy number? > > Please let me know if what I am doing makes sense or if I should be > organizing according to evalue or % identity. > > My other question was what the numbers next to the organism name in > the reference set pull down menu stand for. Many organisms have the > same number, so I was wondering why that is. > > Thanks again for your help. > > Best regards, > Joanne > ________________________________________ > From: Mark DSouza [dsouza(a)mcs.anl.gov] > Sent: August 25, 2011 12:37 PM > To: Joanne Wyglinski, Ms > Cc: mg-rast(a)mcs.anl.gov > Subject: Re: [mg-rast] # reads hit > > Hi, > > The '# reads hit' refers to the number of reads in the metagenomic > sample that were found to have sequence similarity to a gene in the > reference genome, sorting this column in descending order puts the > genes which hit the largest number of reads at the top of the table. > This would give you a sense of the content of the metagenomic dataset > (as compared to the reference genome) while organizing by average > evalue or %identity will highlight other genes based on similarity to > genes in the reference genome. If you let us know what you are looking > for in your metagenomic dataset we will try and provide a more > detailed answer. > > -- > 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 > > > ----- Original Message ----- > > From: "Joanne Wyglinski, Ms" <joanne.wyglinski(a)mail.mcgill.ca> > > To: mg-rast(a)mcs.anl.gov > > Sent: Thursday, August 25, 2011 8:55:50 AM > > Subject: Re: [mg-rast] # reads hit > > Hi, > > > > I have a question regarding metagenomic analysis on your website. > > > > For the organism recruitment plot, the MG-RAST website states that > > you > > can organize according to "# reads hit". > > > > What does this mean? > > > > Also, if I am looking for the most significant data, should I be > > organizing according to "# reads hit" or something like avg evalue > > or > > % identity. > > > > Thanks for your time in advance. Your help is greatly appreciated. > > > > Best regards, > > Joanne Wyglinski

1 0

Re: [mg-rast] MG-Rast uploading problem
by Mark D'Souza 25 Aug '11

25 Aug '11

Hi, Your latest upload worked correctly, the fasta file and quality file are paired and you can now create the job (you could also delete the other two files). Regards, Mark On Aug 25, 2011, at 3:11 PM, Fei Luo wrote: > Hi Mark, > > Yes I have tried that. But still, I cannot extract the qual file... Any > other hint? > > Thanks, > > Fei > > On 11-08-25 3:43 PM, "Mark DSouza" <dsouza(a)mcs.anl.gov> wrote: > >> Hi, >> >> The problem is with the naming of the quality file, you will need to >> rename your files so that they are named as 'polished_trial2.fasta' and >> 'polished_trial2.qual' with a common prefix 'polished_trial2'. The names >> in your zipped file are not getting matched correctly and the qual file >> is getting treated as a fasta file, causing the error. >> >> Regards, >> Mark >> >> ----- Original Message ----- >>> From: "Fei Luo" <fei.luo(a)utoronto.ca> >>> To: "Mark DSouza" <dsouza(a)mcs.anl.gov> >>> Sent: Thursday, August 25, 2011 2:30:53 PM >>> Subject: Re: [mg-rast] MG-Rast uploading problem >>> Hi, >>> >>> Oh sorry I forgot... I was using the account laura.hug >>> >>> Thanks! >>> >>> Fei >>> >>> On 11-08-25 3:25 PM, "Mark DSouza" <dsouza(a)mcs.anl.gov> wrote: >>> >>>> Hi, >>>> >>>> Could you provide us with the username for the account you are using >>>> for >>>> the upload? I did not find a record of uploads for the username >>>> associated with this email address. >>>> >>>> -- >>>> 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 >>>> >>>> >>>> ----- Original Message ----- >>>>> From: "Fei Luo" <fei.luo(a)utoronto.ca> >>>>> To: mg-rast(a)mcs.anl.gov >>>>> Sent: Thursday, August 25, 2011 2:18:43 PM >>>>> Subject: [mg-rast] MG-Rast uploading problem >>>>> Hi, >>>>> >>>>> >>>>> I am trying uploading some assembled contig files to MG-Rast for >>>>> analysis. I was trying uploading a zip file including the fasta >>>>> file >>>>> and fasta.qual file with the same prefix. However, only the fasta >>>>> file >>>>> got extracted and there is always error information for qual file. >>>>> Can >>>>> you give me some suggestion about what was going on and how I can >>>>> get >>>>> it work? >>>>> >>>>> >>>>> Thanks! >>>>> >>>>> >>>>> Fei > >

1 0

Re: [mg-rast] MG-Rast uploading problem
by Mark DSouza 25 Aug '11

25 Aug '11

Hi, The problem is with the naming of the quality file, you will need to rename your files so that they are named as 'polished_trial2.fasta' and 'polished_trial2.qual' with a common prefix 'polished_trial2'. The names in your zipped file are not getting matched correctly and the qual file is getting treated as a fasta file, causing the error. Regards, Mark ----- Original Message ----- > From: "Fei Luo" <fei.luo(a)utoronto.ca> > To: "Mark DSouza" <dsouza(a)mcs.anl.gov> > Sent: Thursday, August 25, 2011 2:30:53 PM > Subject: Re: [mg-rast] MG-Rast uploading problem > Hi, > > Oh sorry I forgot... I was using the account laura.hug > > Thanks! > > Fei > > On 11-08-25 3:25 PM, "Mark DSouza" <dsouza(a)mcs.anl.gov> wrote: > > >Hi, > > > >Could you provide us with the username for the account you are using > >for > >the upload? I did not find a record of uploads for the username > >associated with this email address. > > > >-- > >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 > > > > > >----- Original Message ----- > >> From: "Fei Luo" <fei.luo(a)utoronto.ca> > >> To: mg-rast(a)mcs.anl.gov > >> Sent: Thursday, August 25, 2011 2:18:43 PM > >> Subject: [mg-rast] MG-Rast uploading problem > >> Hi, > >> > >> > >> I am trying uploading some assembled contig files to MG-Rast for > >> analysis. I was trying uploading a zip file including the fasta > >> file > >> and fasta.qual file with the same prefix. However, only the fasta > >> file > >> got extracted and there is always error information for qual file. > >> Can > >> you give me some suggestion about what was going on and how I can > >> get > >> it work? > >> > >> > >> Thanks! > >> > >> > >> Fei

1 0

Re: [mg-rast] MG-Rast uploading problem
by Mark DSouza 25 Aug '11

25 Aug '11

Hi, Could you provide us with the username for the account you are using for the upload? I did not find a record of uploads for the username associated with this email address. -- 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 ----- Original Message ----- > From: "Fei Luo" <fei.luo(a)utoronto.ca> > To: mg-rast(a)mcs.anl.gov > Sent: Thursday, August 25, 2011 2:18:43 PM > Subject: [mg-rast] MG-Rast uploading problem > Hi, > > > I am trying uploading some assembled contig files to MG-Rast for > analysis. I was trying uploading a zip file including the fasta file > and fasta.qual file with the same prefix. However, only the fasta file > got extracted and there is always error information for qual file. Can > you give me some suggestion about what was going on and how I can get > it work? > > > Thanks! > > > Fei

1 0

MG-Rast uploading problem
by Fei Luo 25 Aug '11

25 Aug '11

Hi, I am trying uploading some assembled contig files to MG-Rast for analysis. I was trying uploading a zip file including the fasta file and fasta.qual file with the same prefix. However, only the fasta file got extracted and there is always error information for qual file. Can you give me some suggestion about what was going on and how I can get it work? Thanks! Fei

1 0

Re: [mg-rast] Loading Database : error
by Mark DSouza 25 Aug '11

25 Aug '11

Hi, I'm sorry, the jobs can not be processed because of their small size. The MG-RAST pipeline was designed to efficiently handle large datasets with Gbp of sequence data, unfortunately some of the steps required do not work well for extremely small samples and the processing fails. Your metagenomes with 3, 6 and 23 sequences are considerably below the threshold size required, you may want to consider using the NCBI BLAST website for such a small number of sequences. If you were submitting these as test cases, please feel free to submit much larger datasets, 50-100Mbp would work well to illustrate the systems capabilities. -- 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 ----- Original Message ----- > From: "Ann Womack" <annw(a)uoregon.edu> > To: mg-rast(a)mcs.anl.gov > Sent: Thursday, August 25, 2011 1:12:15 PM > Subject: [mg-rast] Loading Database : error > Hi, > > I'm trying to load 6 very small metagenomes. Two loaded fine > yesterday, but I am repeatedly encountering an error (Loading Database > : error) after uploading the remaining four. The job #s are > 30722-30726 and id's 4471125.3-4471129.3. There are actually five jobs > here because I accidentally uploaded one metagenome twice. Thank you > in advance for any help you can provide. > > Ann > > -- > Ann M. Womack > PhD Candidate > Green Lab > Institute of Ecology and Evolution > University of Oregon > > Office/lab phone: 541-346-5674 > email: annw(a)uoregon.edu

1 0

Loading Database : error
by Ann Womack 25 Aug '11

25 Aug '11

Hi, I'm trying to load 6 very small metagenomes. Two loaded fine yesterday, but I am repeatedly encountering an error (Loading Database : error) after uploading the remaining four. The job #s are 30722-30726 and id's 4471125.3-4471129.3. There are actually five jobs here because I accidentally uploaded one metagenome twice. Thank you in advance for any help you can provide. Ann -- Ann M. Womack PhD Candidate Green Lab Institute of Ecology and Evolution University of Oregon Office/lab phone: 541-346-5674 email: annw(a)uoregon.edu

1 0

Re: [mg-rast] # reads hit
by Mark DSouza 25 Aug '11

25 Aug '11

Hi, The '# reads hit' refers to the number of reads in the metagenomic sample that were found to have sequence similarity to a gene in the reference genome, sorting this column in descending order puts the genes which hit the largest number of reads at the top of the table. This would give you a sense of the content of the metagenomic dataset (as compared to the reference genome) while organizing by average evalue or %identity will highlight other genes based on similarity to genes in the reference genome. If you let us know what you are looking for in your metagenomic dataset we will try and provide a more detailed answer. -- 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 ----- Original Message ----- > From: "Joanne Wyglinski, Ms" <joanne.wyglinski(a)mail.mcgill.ca> > To: mg-rast(a)mcs.anl.gov > Sent: Thursday, August 25, 2011 8:55:50 AM > Subject: Re: [mg-rast] # reads hit > Hi, > > I have a question regarding metagenomic analysis on your website. > > For the organism recruitment plot, the MG-RAST website states that you > can organize according to "# reads hit". > > What does this mean? > > Also, if I am looking for the most significant data, should I be > organizing according to "# reads hit" or something like avg evalue or > % identity. > > Thanks for your time in advance. Your help is greatly appreciated. > > Best regards, > Joanne Wyglinski

1 0