Choosing of selective metacyc pathway"ggpicrust2"

[kesava31]

Hello
I am using ggpicrust2 in R from the PICRUST output files. I am interested to look some of the specific pathways from metacyc output. Below, code I used for plotting heatmap showing the pathways that are significant (< 0.05). I am wondering how to choose selective pathway to make heatmap that are interested for example (Aromatic compound degradation, Pathogenecity).
Could you help me in this regard.. How can I do it with the code. Thanks.

metacyc_daa_results_dfnew <- pathway_daa(abundance = metacyc_abundancenew %>% column_to_rownames("pathway"), metadata = metadata, group = "sample", daa_method = "LinDA")

metacyc_daa_annotated_results_dfnew <- pathway_annotation(pathway = "MetaCyc", daa_results_df = metacyc_daa_results_dfnew, ko_to_kegg = FALSE)

feature_with_p_0.05 <- metacyc_daa_results_dfnew %>% filter(p_adjust < 0.05)
pathway_heatmap(
abundance = metacyc_abundancenew %>%
right_join(
metacyc_daa_annotated_results_dfnew %>% select(all_of(c("feature","description"))),
by = c("pathway" = "feature")
) %>%
filter(pathway %in% feature_with_p_0.05$feature) %>%
select(-"pathway") %>%
column_to_rownames("description"),
metadata = metadata,
group = "sample"
)

[cafferychen777]

Hi there,

Thank you for your question about selecting specific MetaCyc pathways in ggpicrust2. I can help you modify your code to visualize only the pathways you're interested in (like aromatic compound degradation and pathogenicity). Here are several approaches:

1. Using keyword filtering:

# Filter pathways by keywords
interested_pathways <- metacyc_daa_annotated_results_dfnew %>%
  filter(
    # Use regex to match descriptions you're interested in
    grepl("aromatic|degradation|pathogen", description, ignore.case = TRUE)
  )

# Create heatmap with filtered pathways
pathway_heatmap(
  abundance = metacyc_abundancenew %>%
    right_join(
      metacyc_daa_annotated_results_dfnew %>% 
        select(all_of(c("feature","description"))),
      by = c("pathway" = "feature")
    ) %>%
    # Filter to include only pathways of interest
    filter(pathway %in% interested_pathways$feature) %>%
    select(-"pathway") %>%
    column_to_rownames("description"),
  metadata = metadata,
  group = "sample"
)

2. For more precise filtering, you can create a vector of specific keywords:

# Define keywords of interest
keywords <- c(
  "aromatic compound degradation",
  "pathogenicity",
  "virulence",
  # Add more keywords as needed...
)

# Filter using keywords
interested_pathways <- metacyc_daa_annotated_results_dfnew %>%
  filter(
    sapply(keywords, function(x) grepl(x, description, ignore.case = TRUE)) %>%
      apply(1, any)
  )

3. If you know the specific pathway IDs, you can filter directly:

# Use specific pathway IDs
pathway_ids <- c("PWY-6261", "PWY-6126") # Replace with your pathways of interest

interested_pathways <- metacyc_daa_annotated_results_dfnew %>%
  filter(feature %in% pathway_ids)

Tips:

• You can first view all pathway descriptions using View(metacyc_daa_annotated_results_dfnew) to help identify the ones you're interested in
• The ignore.case = TRUE parameter ensures case-insensitive matching
• Adjust the keywords or filtering criteria based on your specific needs
• You can combine these methods as needed

Let me know if you need any clarification or have additional questions!