FileNotFoundError: [Errno 2] No such file or directory: './Data/ReichLabEigenstrat/Raw/meta.csv'

[mariamnawaz1]

Hi,

I am combing results from hapsb_ind() output using the pp_individual_roh() and I don't want to add meta information. But pp_individual_roh() throws me an error even if I do meta_info=False. It seems to me it is because it is reading meta_path before checking meta_info which happens only during the merging step. Am I missing something?

Function:

def pp_individual_roh(iids, meta_path="./Data/ReichLabEigenstrat/Raw/meta.csv", 
                      base_folder="./Empirical/Eigenstrat/Reichall/", 
                      suffix='_roh_full.csv', save_path="", min_cm=[4,8,12], snp_cm=50, 
                      gap=0.5, min_len1=2.0, min_len2=4.0,
                      output=True, meta_info=True):
    """Post-process Individual ROH .csv files. Combines them into one summary ROH.csv, saved in save_path.
    Use Individuals iids, create paths and run the combining.
    iids: List of target Individuals
    base_folder: Folder where to find individual results .csvs
    min_cm: Minimum post-processed Length of ROH blocks. Array (to have multiple possible values)
    snp_cm: Minimum Number of SNPs per cM
    gap: Maximum length of gaps to merge
    output: Whether to plot output per Individual.
    meta_info: Whether to merge in Meta-Info from the original Meta File
    save_path: If given, save resulting dataframe there
    min_len1: Minimum Length of shorter block to merge [cM]
    min_len2: Maximum Length of longer block to merge [cM]"""
    
    ### Look up Individuals in meta_df and extract relevant sub-table
    df_full = pd.read_csv(meta_path)
    df_meta = df_full[df_full["iid"].isin(iids)]  # Extract only relevant Indivdiuals
    
    print(f"Loaded {len(df_meta)} / {len(df_full)} Individuals from Meta")
    
    paths = give_iid_paths(df_meta["iid"], base_folder=base_folder, suffix=suffix)
    df1 = create_combined_ROH_df(paths, df_meta["iid"].values, df_meta['clst'].values, 
                                 min_cm=min_cm, snp_cm=snp_cm, gap=gap, 
                                 min_len1=min_len1, min_len2=min_len2, output=output)
    
    ### Merge results with Meta-Dataframe
    if meta_info:
        df1 = pd.merge(df1, df_meta, on="iid")
        
    if len(save_path) > 0:
        df1.to_csv(save_path, sep="\t", index=False)
        print(f"Saved to: {save_path}")
    
    return df1

[hringbauer]

Ah yes, this function for post-processing is overly complicated as it strictly requires a meta table path.

"meta_info" just indicates whether the info from the meta file should be part of the output table - but it does not turn off the requirement to provide such a meta table.

As a quick fix, you can simply create a table with "iid" and "clst" columns - I know that this is not ideal.

I will try to update this function to make the meta table file fully optional. Stay tuned for the next release!