pathway_daa(): There are no statistically significant biomarkers

[GeoffreyOlweny]

What could be the meaning of this error. May metadata file seems to be okay but this is the result I get when I run the ggpicrust() func.

[GeoffreyOlweny]

Here is a snippet of my code; metadata <-

• read_delim(
• "C:/Users/hp/Desktop/COPD_analysis/new classification/Extracted/OMV_COPD_metadata_Dada2.txt",
• delim = "\t",
• escape_double = FALSE, trim_ws = TRUE) kegg_abundance <-
• ko2kegg_abundance(
• "C:/Users/hp/Desktop/Picrust_COPD/picrust input/picrust2_out_pipeline/KO_metagenome_out/pred_metagenome_unstrat.tsv") daa_results_list <-
• ggpicrust2(
• file = "C:/Users/hp/Desktop/Picrust_COPD/picrust input/picrust2_out_pipeline/KO_metagenome_out/pred_metagenome_unstrat.tsv",
• metadata = metadata,
• group = "cohort",
• pathway = "KO",
• daa_method = "LinDA",
• p_values_bar = TRUE,
• p.adjust = "BH",
• ko_to_kegg = TRUE,
• order = "cohort",
• select = NULL,
• reference = "1" The kegg pathway with zero abundance in all the different samples has been removed. 0 features are filtered! The filtered data has 76 samples and 225 features will be tested! Warning: Some features have less than 3 nonzero values! They have virtually no statistical power. You may consider filtering them in the analysis! Imputation approach is used. Fit linear models ... Completed. Error in ggpicrust2(file = "C:/Users/hp/Desktop/Picrust_COPD/picrust input/picrust2_out_pipeline/KO_metagenome_out/pred_metagenome_unstrat.tsv", : There are no statistically significant biomarkers

Note: I get the same error when I follow the step-by-step process on your tutorial. Attached is my metadata file and picrust2 abundance output file. Thank you very much for your insightful picrust2 output downstream analysis package ggpicrust2_input.zip

[cafferychen777]

Hi there,

Thank you for reaching out. Based on the code snippet you provided, I can see that the error message you received is not actually an error, but rather a result of the statistical test performed by the ggpicrust2 package. The message indicates that there is no statistically significant difference in the pathway abundance between the two groups you are comparing.

Therefore, this means that there is no evidence of differential signal pathway activity between the groups. Please note that the warning message about features with less than 3 nonzero values also suggests that some of the features may have very little statistical power and may need to be filtered in your analysis.

I hope this clarifies things for you. If you have any further questions or concerns, please don't hesitate to ask.

Best regards.

[GeoffreyOlweny]

Does this mean that I can not proceed with the downstream visualizations in ggpicrust2?

OLWENY GEOFFREY LAB TECHNOLOGIST. MYCOBACTERIOLOGY LABORATORY COLLEGE OF HEALTH SCIENCES. MAKERERE UNIVERSITY. TEL:0788655538

On Sun, May 7, 2023 at 3:37 PM Caffery Yang @.***> wrote:

Hi there,

Thank you for reaching out. Based on the code snippet you provided, I can see that the error message you received is not actually an error, but rather a result of the statistical test performed by the ggpicrust2 package. The message indicates that there is no statistically significant difference in the pathway abundance between the two groups you are comparing.

Therefore, this means that there is no evidence of differential signal pathway activity between the groups. Please note that the warning message about features with less than 3 nonzero values also suggests that some of the features may have very little statistical power and may need to be filtered in your analysis.

I hope this clarifies things for you. If you have any further questions or concerns, please don't hesitate to ask.

Best regards.

— Reply to this email directly, view it on GitHub https://github.com/cafferychen777/ggpicrust2/issues/23#issuecomment-1537431524, or unsubscribe https://github.com/notifications/unsubscribe-auth/AXTMMUG7SRQUMQDSUXZ3W2TXE6JRTANCNFSM6AAAAAAXYXD4FQ . You are receiving this because you authored the thread.Message ID: @.***>

[cafferychen777]

You can try other differential abundance methods to see if there are different biomarker with statistically significance.