Where can I find example files to try MEGAN?


#1

A wide selection of different metagenomic datasets are provided on the public MEGAN server.
This datasets include for example analysis of permafrost samples, human gut microbiomes, the HMP mock datasets and more.

To connect to the public MEGANServer in MEGAN, select File -> Open from Server… This will open the Remote Connection Dialog.

The public server is found at TODO, and the username and password are both guest
After filling out the information, your Window should look like this:

By clicking on Open, the connection to the server is established, and you can now browse the directory system including all the example datasets.

Feel free to analyse, visualise and adapt the examples to your liking, as files from MEGANServer are opened in read-only mode, so you are not risking to loose any information by doing something wrong.


#2

Currently available datasets on the public MeganServer:

permafrost (2 cores, each with 2 layers sequenced at 3 different timepoints)
summer-school (Datasets from the BEST Summer School 2015, including the even mock community data from HMP as well as one human gut microbiome dataset)


#3

Hi All -
The permafrost .rma6 datasets in the MeganServer work great and have been a lot of fun to look through. But I am struggling to format my input files correctly to import into Megan to create my own datasets. It would be very helpful to look at the raw files you imported to create the .rma6 datasets. Can you provide those? My files are metagenome annotations downloaded from the Joint Genome Institute IMG system. I want to reformat the files so that I can open them in MEGAN. Can you help?


#4

Could you please make a file of interest containing “IMG metagenome annotations” available to me and I will take a look at this.


#5

I’ve taken a look through the example files but I wasn’t able to see any of the original files that were originally imported into MEGAN. It would be helpful to see what those look like, especially the format from the RDP stand alone classifier.

I’m currently trying to import assignments from the RDP classifier - stand alone version and I am having problems. In fact, files that I imported without problems 2 years ago using MEGAN v 5 won’t open using MEGAN6. Files from the online classifier open just fine. I’ve included a portion of my current RDPstandalone classifier results that I can’t seem to import into MEGAN6.

test.out (914 Bytes)


#6

I took a look at your file. This looks quite different to what MEGAN expects as “RDP standalone format”, which looks like this:

> Sequence1
Root; 1.0; Bacteria; 1.0; Proteobacteria; 1.0; Alphaproteobacteria; 1.0; Rhizobiales; 1.0; Hyphomicrobiaceae; 0.84; Filomicrobium; 0.8;
> Sequence2
Root; 1.0; Bacteria; 1.0; Proteobacteria; 1.0; Alphaproteobacteria; 1.0; Rhodobacterales; 0.52; Rhodobacteraceae; 0.52; Pannonibacter; 0.4;
> Sequence3
Root; 1.0; Bacteria; 1.0; Proteobacteria; 1.0; Gammaproteobacteria; 1.0; Legionellales; 0.51; Coxiellaceae; 0.51; Rickettsiella; 0.51;

So it looks like the output format of the RDP standalone classifier has changed. We will look into this and will modify MEGAN accordingly.


#7

Would it be possible to add .csv example files for each of the different Classifications (KEGG; COG; …) on the public MEGAN server?
I tried my own files which worked in the previous version (MEGAN5) but don`t seem to work with the new version, as all information is imported but not identified and categorized as “not assigned”.


#8

Is the taxonomic information assigned and the functional is not or all reads are generally not assigned to anything?

Can you please post some lines of the .csv files you are trying to import and which mapping files and MEGAN6 version you are using?

On the public server there are multiple samples which include functional annotation which you can export as .csv
For example, opening the ASARI -> Alice00.daa, showing COG annotation and selecting only the Level 1 nodes gave me this very basic .csv file (selecting eggnogname_to_count)

“informationStorageAndProcessing” 1740827.0
"cellularProcessesAndSignaling" 1584097.0
"metabolism" 3443847.0


#9

Thanks for the fast reply.

I am only using a file where reads were assigned to function (KEGG). The import of csv files with taxonomic information works perfectly.

This is how my .csv file looks like. I am using the “Community Edition”.

I also tried to export data from an example to a .csv file, but this seem to not work for lower levels. In fact I can`t export data as .csv file if I, for example want only the reads assigned to “Glycolysis”.

ko_id OM1 OM2 OM3
K00001 77 164 659
K00002 0 0 3
K00003 153 493 2626
K00004 2 9 72


#10

This is probably because KEGG mapping was improved in MEGAN6.
The first lines of all leaves below KO00010 (glycolysis/gluconeogenesis) as a csv now look something like this:

“K00001 alcohol dehydrogenase [EC:1.1.1.1]” 1670.0
"K00002 alcohol dehydrogenase (NADP+) [EC:1.1.1.2]" 111.0
"K00016 L-lactate dehydrogenase [EC:1.1.1.27]" 3198.0