merge_OTU_tsv_files

Goal: merge taxonomy information output as .tsv files from different taxonomy classifiers into an Excel sheet, ultimately allowing comparison of concordant vs. discordant taxon classifications between different programs

Initial file formats

• We are starting with an Excel file with statistically significant hits from an experiment, listed as OTU i and OTU j (analysis involves interactions between bacterial pairs within a system)
• Each OTU i and j has been assigned a taxonomy classifier (Taxon i and Taxon j, respectively) by greengenes 1 (?)
• We have two .tsv files that include taxonomy information obtained by (1) SILVA and (2) greengenes 2

Changes I made because I didn't know how to do this in R

• Change OTU i and OTU j in Excel sheet to Feature_ID_i and Feature_ID_j
• Copy .tsv to .xlsx file
• Change Feature.ID to Feature_ID_i and save as (name of .tsv file)_i
• Made a copy and changed Feature_ID_i to Feature_ID_j and save as (name of .tsv file)_j

Commands learned in this exercise

• read in excel file

    df <- read_excel()
  

• convert features in the df from doubles/strings to characters

    df <- df %>%
    mutate(columnheader = as.character(columnheader))
  

• merge dfs

    df_merged <- df %>%
    left_join(df_2, by = c("Feature_ID" = "Feature_ID")
  

Ways to try to optimize

• Write function to repeat greengenes2 actions with Silva