From 5fe385990c79068bf68d746c2af6d6308b6bb543 Mon Sep 17 00:00:00 2001 From: zqfang Date: Fri, 23 Aug 2024 11:43:12 -0700 Subject: [PATCH] compatible with pandas >2.1 --- gseapy/gsea.py | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/gseapy/gsea.py b/gseapy/gsea.py index 9cd0009..3105702 100644 --- a/gseapy/gsea.py +++ b/gseapy/gsea.py @@ -107,7 +107,11 @@ def _filter_data(self, df: pd.DataFrame) -> pd.DataFrame: # fix numeric_only error df_std = df.groupby(by=cls_dict, axis=1).std(ddof=0) else: - df_std = df.groupby(by=cls_dict, axis=1).std(numeric_only=True, ddof=0) + gene_idxs = df.index.to_list() + # df_std = df.groupby(by=cls_dict, axis=1).std(numeric_only=True, ddof=0) + df_std = ( + df.T.groupby(by=cls_dict)[gene_idxs].std(numeric_only=True, ddof=0).T + ) # remove rows that are all zeros ! df = df.loc[df.abs().sum(axis=1) > 0, :] @@ -180,8 +184,11 @@ def calc_metric( df_mean = df.groupby(by=classes, axis=1).mean() df_std = df.groupby(by=classes, axis=1).std() else: - df_mean = df.groupby(by=classes, axis=1).mean(numeric_only=True) - df_std = df.groupby(by=classes, axis=1).std(numeric_only=True) + gene_idxs = df.index.to_list() + df_std = df.T.groupby(by=classes)[gene_idxs].std(numeric_only=True).T + df_mean = df.T.groupby(by=classes)[gene_idxs].mean(numeric_only=True).T + # df_mean = df.groupby(by=classes, axis=1).mean(numeric_only=True) + # df_std = df.groupby(by=classes, axis=1).std(numeric_only=True) class_values = Counter(classes.values()) n_pos = class_values[pos] n_neg = class_values[neg]