Metagenome prediction using ggpicrust2

[Anandi Batabyal]

Hello friends,

I am trying to perform metagenome prediction based on my picrust2-generated output files. Used this code:

picrust2_pipeline.py \
  -s cleaned-dna-sequences.fasta \
  -i feature-table.biom \
  -o picrust2_out_pipeline/picrust2_output \
  -p 8 

I got EC_metagenome_out, KO_metagenome_out and pathways_out output folders. 

In KO_metagenome_out and EC_metagenome out folders each, I have pred_metagenome_unstrat.tsv.gz, seqtab_norm.tsv.gz and weighted_nsti.tsv.gz files. 

In my pathways_out folder, I have path_abun_unstrat.tsv.gz file. 

I want to visualize specific pathways that are differentially enriched or depleted in my treatment versus control group. I used pred_ko_metagenome_unstrat.tsv file for functional prediction. I used kegg_abundance <- ggpicrust2::ko2kegg_abundance(filename) followed by daa_results_df <- pathway_daa(abundance = kegg_abundance, metadata = metadata, group = "XYZ", daa_method = "ALDEx2", select = NULL, p.adjust = "BH", reference = "X") followed by daa_annotated_results_df <- pathway_annotation(

  pathway = "KO",
  daa_results_df = daa_results_df,  
  ko_to_kegg = TRUE
) 

followed by 

pathway_errorbar(
  abundance = ko_abundance,  
  daa_results_df = daa_annotated_results_df,
  Group = matching_group,  
  p_values_threshold = 0.1,
  order = "p_values",  
  select = selected_features,  
  ko_to_kegg = FALSE,
  p_value_bar = TRUE,                                            
  colors = NULL,
  x_lab = "pathway_name"
)

Is this the right approach? I am not sure what are the differences among the different files generated from picrust2 and which one I should use for visualization to achieve my desired output. The pathway_errorbar function is throwing error and I am not able to get the visualization. Is there a better way to do all this?

I am a beginner at using PICRUST2, and any help will be greatly appreciated.

Thanks,

Anandi

[Robyn Wright]

Hi Anandi,

I can't answer questions about ggpicrust2 as that has different creators - I think they have sometimes answered questions here, but you'd be best going to their Github page with questions for them. 

If you are interested in visualising pathways then you would need to use the files in the pathways_out folder. The different folders correspond to different methods of grouping/annotating functions. There are lots of different functional databases and these all overlap to various extents. Some of these are better or worse for particular functions, and they group at different levels of resolution. The pathway prediction is performed based on the EC (Enzyme Commission) results - you can see some more information about the pathway predictions in the FAQ. The tutorial also provides more information about what exactly is happening at each step of the PICRUSt2 pipeline. 

Best wishes,

Robyn

[Kajal Kumari]

I have two raw data of fastq file ...how to make biom file and fasta file to run in picrust2

--
You received this message because you are subscribed to the Google Groups "picrust-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to picrust-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/picrust-users/28ecc394-7043-4591-ae7d-ae5093dd540cn%40googlegroups.com.

[Robyn Wright]

Hi there, 

Before you can use PICRUSt2 you first need to process your raw data to generate representative sequences and a feature table. Then you can use PICRUSt2 to predict functions within your data. I recommend using QIIME2, which has lots of useful tutorials on their website as well as a forum.

Robyn

[Kajal Kumari]

Can't install  qiime

To view this discussion visit https://groups.google.com/d/msgid/picrust-users/5d947924-cc8e-4abe-b606-a82fad823055n%40googlegroups.com.

[Kajal Kumari]

Please give commands bcz I try many times to install qiime...

[Robyn Wright]

Hi there,

You can find comprehensive documentation and tutorials on the QIIME2 website: https://amplicon-docs.readthedocs.io/en/latest/

This includes information on how to install QIIME2: https://amplicon-docs.readthedocs.io/en/latest/how-to-guides/install.html

If you are having issues with your installation and have specific error messages that you can contact the QIIME2 team about, I recommend posting on their forum. 

Robyn

[Kajal Kumari]

How to contact qiime2 team?

To view this discussion visit https://groups.google.com/d/msgid/picrust-users/229ba277-e04c-4c54-8fdc-e6dae7e6377cn%40googlegroups.com.

[Robyn Wright]

I already said - on their forum. 

[Kajal Kumari]

Help

To view this discussion visit https://groups.google.com/d/msgid/picrust-users/6f254757-f5cb-4c34-a163-98afe6f52111n%40googlegroups.com.