From 222255684163afce821d3cfb3edfd812678b7795 Mon Sep 17 00:00:00 2001
From: Daniel Bojar <daniel@bojar.net>
Date: Wed, 17 Apr 2024 07:20:21 +0200
Subject: [PATCH] Nothing transform for comparison; make min_samples more
 robust

---
 build/lib/glycowork/motif/analysis.py | 2 ++
 glycowork/glycan_data/stats.py        | 6 +++---
 glycowork/motif/analysis.py           | 2 ++
 3 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/build/lib/glycowork/motif/analysis.py b/build/lib/glycowork/motif/analysis.py
index 502fa05..350f716 100644
--- a/build/lib/glycowork/motif/analysis.py
+++ b/build/lib/glycowork/motif/analysis.py
@@ -566,6 +566,8 @@ def get_differential_expression(df, group1, group2,
     df = get_additive_logratio_transformation(df, group1, group2, paired = paired)
   elif transform == "CLR":
     df.iloc[:, 1:] = clr_transformation(df.iloc[:, 1:], group1, group2, gamma = gamma, custom_scale = custom_scale)
+  elif transform == "Nothing":
+    pass
   else:
     raise ValueError("Only ALR and CLR are valid transforms for now.")
   # Sample-size aware alpha via Bayesian-Adaptive Alpha Adjustment
diff --git a/glycowork/glycan_data/stats.py b/glycowork/glycan_data/stats.py
index 760d221..74542d7 100644
--- a/glycowork/glycan_data/stats.py
+++ b/glycowork/glycan_data/stats.py
@@ -234,9 +234,9 @@ def impute_and_normalize(df, groups, impute = True, min_samples = 0.1):
     | Returns a dataframe in the same style as the input 
     """
     if min_samples:
-      min_count = np.floor(df.shape[1] * min_samples)
+      min_count = max(np.floor(df.shape[1] * min_samples), 2) + 1
       mask = (df != 0).sum(axis = 1) >= min_count
-      df = df[mask]
+      df = df[mask].reset_index(drop = True)
     colname = df.columns[0]
     glycans = df[colname]
     df = df.iloc[:, 1:]
@@ -998,7 +998,7 @@ def correct_multiple_testing(pvals, alpha):
   corrpvals = [p if p >= pvals[i] else pvals[i] for i, p in enumerate(corrpvals)]
   significance = [p < alpha for p in corrpvals]
   if sum(significance) > 0.9*len(significance):
-    print("Significance inflation detected. The CLR/ALR transformation cannot seem to handle this dataset.\
+    print("Significance inflation detected. The CLR/ALR transformation cannot seem to handle this dataset. Consider running again with a higher gamma value.\
              Proceed with caution; for now switching to Bonferroni correction to be conservative about this.")
     res = multipletests(pvals, method = 'bonferroni')
     corrpvals, alpha = res[1], res[3]
diff --git a/glycowork/motif/analysis.py b/glycowork/motif/analysis.py
index 502fa05..350f716 100644
--- a/glycowork/motif/analysis.py
+++ b/glycowork/motif/analysis.py
@@ -566,6 +566,8 @@ def get_differential_expression(df, group1, group2,
     df = get_additive_logratio_transformation(df, group1, group2, paired = paired)
   elif transform == "CLR":
     df.iloc[:, 1:] = clr_transformation(df.iloc[:, 1:], group1, group2, gamma = gamma, custom_scale = custom_scale)
+  elif transform == "Nothing":
+    pass
   else:
     raise ValueError("Only ALR and CLR are valid transforms for now.")
   # Sample-size aware alpha via Bayesian-Adaptive Alpha Adjustment