diff --git a/libworkflows/libworkflows/libworkflows.py b/libworkflows/libworkflows/libworkflows.py index 0a89e12b64ef0d48dcdfb9dbcead5c02046f3b34..849ceacff06511fcee2481529f5ff4b9f4fa3c7e 100644 --- a/libworkflows/libworkflows/libworkflows.py +++ b/libworkflows/libworkflows/libworkflows.py @@ -245,7 +245,7 @@ def sum_htseq_counts(counts_filename): def read_htseq_counts(counts_filename): - return pd.read_table(counts_filename, header=None, index_col=0).drop( + return pd.read_csv(counts_filename, sep="\t", header=None, index_col=0).drop( ["__no_feature", "__ambiguous", "__too_low_aQual", @@ -260,11 +260,11 @@ def sum_feature_counts(counts_filename, nb_bams=1): This determines which columns should be used.""" # Counts are in the 7-th column, starting from third row. # The first sum is over the the rows, the second over the columns - return pd.read_table(counts_filename, skiprows=2, usecols=range(6, 6 + nb_bams), header=None).sum().sum() + return pd.read_csv(counts_filename, sep="\t", skiprows=2, usecols=range(6, 6 + nb_bams), header=None).sum().sum() def read_feature_counts(counts_filename, nb_bams=1): - return pd.read_table(counts_filename, skiprows=1, usecols=[0, *range(6, 6 + nb_bams)], index_col=0) + return pd.read_csv(counts_filename, sep="\t", skiprows=1, usecols=[0, *range(6, 6 + nb_bams)], index_col=0) # I 3746 3909 "WBGene00023193" - . 17996 @@ -281,7 +281,7 @@ def sum_intersect_counts(counts_filename): """Sums all counts in a bedtools intersect generated *counts_filename*, where the annotation was in bed format.""" # Counts are in the 7-th column try: - return pd.read_table(counts_filename, usecols=[6], header=None).sum().iloc[0] + return pd.read_csv(counts_filename, sep="\t", usecols=[6], header=None).sum().iloc[0] except pd.errors.EmptyDataError: return "NA" @@ -289,7 +289,7 @@ def read_intersect_counts(counts_filename): # index_col takes effect after column selection with usecols, hence index_col=0 (ex-third column): # https://stackoverflow.com/a/45943627/1878788 try: - return pd.read_table(counts_filename, usecols=[3,6], header=None, index_col=0) + return pd.read_csv(counts_filename, sep="\t", usecols=[3,6], header=None, index_col=0) except pd.errors.EmptyDataError: return pd.DataFrame(index = [], columns = ["gene", "counts"]).set_index("gene")