kraken_file error when running MetaPhage (docker)

[vanzyllj]

kraken_file error when running MetaPhage (docker)

[vanzyllj]

I'm using the Docker image btw.

[telatin]

Hi, thanks for reporting. Are you using the 0.3 release or did you clone the repository?

[vanzyllj]

I cloned into the repository

[telatin]

For
The records, I just tested the whole pipeline (both v0.3 and last commit) using docker and worked. Hence the issue is probably related to your specific dataset/metadata file. I will check a way to gather more info

[telatin]

Did you create a "configuration file" using the provided "newProject.py" script?
See https://mattiapandolfovr.github.io/MetaPhage/new

Can you please upload the configuration file used?

[vanzyllj]

Dear Mattia, i've managed to get the program to run using the docker image, but now it gets stuck at "violin_plots" stage with the error: ``` Error executing process > 'violin_plots (megahit)' Caused by: Process `violin_plots (megahit)` terminated with an error exit status (1) Command executed: Rscript /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/bin/Rscript/violin_plot.R count_table.csv taxonomy_table.csv metadata.csv library Command exit status: 1 Command output: (empty) Command error: Error in validObject(.Object) : invalid class “phyloseq” object: Component sample names do not match. Try sample_names() Calls: phyloseq ... do.call -> new -> initialize -> initialize -> validObject Execution halted Work dir: /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/work/ed/c65aacbf9a9ffc654e64e9665a9017 Tip: you can replicate the issue by changing to the process work dir and entering the command `bash .command.run` ``` I've attached my metadata file and config file which was initially created with the python script, but later manually edited for the machine i'm running on (Ubuntu 20.04, 16 core I7 11900, 128gb RAM). The metadata file might be an issue (phyloseq object; issue with rows or columns of the count_table versus taxonomy_table versus metadata.csv). Regards Lonnie

…

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

[MattiaPandolfoVR]

Hi Lonnie, could you please provide your metadata and config file? I fail to find them in the mail attachments.

[vanzyllj]

Dear Mattia, I've attached them again (metadat.csv and nextflow.conf). If you don't see them, I'll upload to Google drive and share the folder with you. Regards Lonnie

…

On Wed, May 25, 2022 at 11:28 AM Mattia Pandolfo ***@***.***> wrote: Hi Lonnie, could you please provide your metadata and config file? I fail to find them in the mail attachments. — Reply to this email directly, view it on GitHub <#44 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZDP2WJWWWM3JFHXYVGU2EDVLXXELANCNFSM5VSR3QHQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

[MattiaPandolfoVR]

Hi again,

for some reason the mail has no attachments. Upload the files on Google drive and send me the folder link, thank you.

Regards
Mattia

[telatin]

See https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/attaching-files to upload directly on github issue (not via mail).
Cheers
a

[vanzyllj]

Here are the config and metadata files:

metadata.csv
nextflow.txt

[MattiaPandolfoVR]

Hi Lonnie,

your Sample names in the metadata.csv are numbers, which cause the sample_data() function to add the "sa" string in front. This function is used to create the phyloseq object, therefore the error: phyloseq tries to match the sample names in your count table (1 2 3 4) to these modified names (sa1, sa2, sa3, sa4), failing. I would suggest you to rename the samples adding some characters (e.g. sample1, sample2 etc).

Regards
Mattia

[vanzyllj]

Great, thank you Mattia, I appreciate your help in sorting this out. I'll let you know if this sorts the issue. Regards Lonnie

…

On Wed, May 25, 2022 at 1:17 PM Mattia Pandolfo ***@***.***> wrote: Hi Lonnie, your Sample names in the metadata.csv are numbers, which cause the sample_data() function to add the "sa" string in front. This function is used to create the phyloseq object, therefore the error: phyloseq tries to match the sample names in your count table (1 2 3 4) to these modified names (sa1, sa2, sa3, sa4), failing. I would suggest you to rename the samples adding some characters (e.g. sample1, sample2 etc). Regards Mattia — Reply to this email directly, view it on GitHub <#44 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZDP2WMGKBLJGN6JWXS4BCDVLYD47ANCNFSM5VSR3QHQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

[vanzyllj]

Hi Mattia,

Unfortunately the change to the metadata file did not resolve the issue. Phyloseq still won't run. Please see the error below:

[a5/df1fca] process > summary (megahit) [100%] 1 of 1 ✔
[c1/fac0fa] process > file_chopper (megahit) [100%] 1 of 1 ✔
[- ] process > taxonomy_table (megahit) -
[30/f61abd] process > heatmap (megahit) [100%] 1 of 1, failed: 1 ✘
[- ] process > alpha_diversity [ 0%] 0 of 1
[- ] process > betadiversity [ 0%] 0 of 1
[- ] process > violin_plots [ 0%] 0 of 1
[- ] process > multiqc -
Error executing process > 'heatmap (megahit)'

Caused by:
Process heatmap (megahit) terminated with an error exit status (1)

Command executed:

Rscript /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/bin/Rscript/heatmap.R count_table.csv taxonomy_table.csv metadata.csv library

Command exit status:
1

Command output:
(empty)

Command error:
Error in validObject(.Object) : invalid class “otu_table” object:
OTU abundance data must have non-zero dimensions.
Calls: filter_taxa ... .nextMethod -> callNextMethod -> .nextMethod -> validObject
Execution halted

Work dir:
/home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/work/30/f61abd786bdda3c28b76651a0f8d80

Tip: you can replicate the issue by changing to the process work dir and entering the command bash .command.run

Here are the new metadata and config files as well as the taxonomy and count tables:

metadata.csv

[nextflow.txt](https://github.com/MattiaPandolfoVR/MetaPhage/files/

count_table.csv

taxonomy_table.csv

[vanzyllj]

nextflow.txt

[vanzyllj]

Hi Mattia, Not sure if you saw my last post. Phyoseq still does not seem to be working (formatting issue with the metadata, count or taxonomy files). Have you had any issues reported relating to the OS? I'm on Ubuntu 20.04. Might try installing it on 18.04 and see if it works there. Regards Lonnie

…

On Wed, May 25, 2022 at 1:25 PM Lonnie van zyl ***@***.***> wrote: Great, thank you Mattia, I appreciate your help in sorting this out. I'll let you know if this sorts the issue. Regards Lonnie On Wed, May 25, 2022 at 1:17 PM Mattia Pandolfo ***@***.***> wrote: > Hi Lonnie, > > your Sample names in the metadata.csv are numbers, which cause the > sample_data() function to add the "sa" string in front. This function is > used to create the phyloseq object, therefore the error: phyloseq tries to > match the sample names in your count table (1 2 3 4) to these modified > names (sa1, sa2, sa3, sa4), failing. I would suggest you to rename the > samples adding some characters (e.g. sample1, sample2 etc). > > Regards > Mattia > > — > Reply to this email directly, view it on GitHub > <#44 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AZDP2WMGKBLJGN6JWXS4BCDVLYD47ANCNFSM5VSR3QHQ> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > -- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

[MattiaPandolfoVR]

Hi Lonnie, sorry for the late reply,

No need to install different distros, the problem was related to the relative abundance filter applied to the phyloseq object, which i fixed. I already updated the code, download again the bin/Rscript folder only and launch MetaPhage (use -resume when you launch it to start from where it stopped). I hope this fix the issue!

Regards
Mattia

[vanzyllj]

Hi Mattia, I eventually came to the same conclusion looking at the R script, but I don't know what I'm doing when it comes to scripting. Glad the problem was resolved. Thanks for your help and persistence. Keep well. Lonnie

…

On Tue, May 31, 2022 at 9:49 AM Mattia Pandolfo ***@***.***> wrote: Hi Lonnie, sorry for the late reply, No need to install different distros, the problem was related to the relative abundance filter applied to the phyloseq object, which i fixed. I already updated the code, download again the bin/Rscript folder only and launch MetaPhage (use -resume when you launch it to start from where it stopped). I hope this fix the issue! Regards Mattia — Reply to this email directly, view it on GitHub <#44 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZDP2WNYKCZYSHQRCTHXDVLVMXAB3ANCNFSM5VSR3QHQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa

[vanzyllj]

Hi Mattia,

alpha diversity runs fine, however the heatmap and beta diversity still have issues.

The beta diversityy error is as follows:

Error executing process > 'betadiversity (megahit)'

Caused by:
Process betadiversity (megahit) terminated with an error exit status (1)

Command executed:

Rscript /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/bin/Rscript/beta_diversity.R count_table.csv taxonomy_table.csv metadata.csv Site

Command exit status:
1

Command output:
(empty)

Command error:
Error in eval(expr, data, expr_env) : object 'Axis.2' not found
Calls: ... -> -> eval_expr -> eval -> eval
Execution halted

Whereas the heatmap error is:

Error executing process > 'heatmap (megahit)'

Caused by:
Process heatmap (megahit) terminated with an error exit status (1)

Command executed:

Rscript /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/bin/Rscript/heatmap.R count_table.csv taxonomy_table.csv metadata.csv Site

Command exit status:
1

Command output:
(empty)

Command error:
Error in cluster::pam(sdata, k, diss = diss, ...) :
Number of clusters 'k' must be in {1,2, .., n-1}; hence n >= 2
Calls: %>% ... heatmaply.default -> heatmapr -> find_k -> pamk ->
Execution halted

The violin_plots error is:

Error executing process > 'violin_plots (megahit)'

Caused by:
Process violin_plots (megahit) terminated with an error exit status (1)

Command executed:

Rscript /home/imbm-bioinformatics/Downloads/MetaPhage-0.3.0/bin/Rscript/violin_plot.R count_table.csv taxonomy_table.csv metadata.csv Site

Command exit status:
1

Command output:
(empty)

Command error:
Error in file(filename, "r", encoding = encoding) :
cannot open the connection
Calls: shhh -> withCallingHandlers -> source -> file
In addition: Warning message:
In file(filename, "r", encoding = encoding) :
cannot open file 'bin/Rscript/filter&CSSnormalize.R': No such file or directory
Execution halted

Regards
Lonnie

[MattiaPandolfoVR]

Hi Lonnie,

Your problem may be related to that "cannot open file 'bin/Rscript/filter&CSSnormalize.R': No such file or directory".
Are you sure you downloaded again the /bin/Rscript folder?

Regards,
Mattia

[vanzyllj]

Hi Mattia,

Yes, that is the issue for violin plots, and it's easy enough to fix by editing the R.script file, but the other two are my main concern.

Regards
Lonnie

[MattiaPandolfoVR]

Hi there,

The problem was related to the sourcing of the filtering function. I updated the Rscripts using it, and tested (again) everything. It seems to work flawless, so download again the whole bin/Rscript folder to have the updated version. Let me know if the issue persist.

Regards,
Mattia

[vanzyllj]

Hi Mattia, I'll really appreciate your efforts in fixing these issues !! I'll give it a go and let you know. Enjoy the evening. Cheers Lonnie

…

On Wed, Jun 1, 2022 at 6:23 PM Mattia Pandolfo ***@***.***> wrote: Hi there, The problem was related to the sourcing of the filtering function. I updated the Rscripts using it, and tested (again) everything. It seems to work flawless, so download again the whole bin/Rscript folder to have the updated version. Let me know if the issue persist. Regards, Mattia — Reply to this email directly, view it on GitHub <#44 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZDP2WLRFYOHOTNZ7BUCD5TVM6E7VANCNFSM5VSR3QHQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Dr. Lonnie van Zyl Chief Officer Institute for Microbial Biotechnology and Metagenomics (IMBM) University of the Western Cape Bellville South Africa