Merge pull request #49 from yangchen2/master

PR to address issues #48, #50, and #51

README.md

@@ -103,4 +103,4 @@ See the [tutorial](tutorial.md) page for a walkthroughon using Qadabra workflow
 Coming soon: An [FAQs](FAQs.md) page of commonly asked question on the statistics and code pertaining to Qadabra.
 
 ## Citation
-The manuscript for Qadabra is currently in progress. Please cite this GitHub page if Qadabra is used for your analysis. This project is licensed under the MIT License. See the [license](LICENSE) file for details.
+The manuscript for Qadabra is currently in progress. Please cite this GitHub page if Qadabra is used for your analysis. This project is licensed under the BSD-3 License. See the [license](LICENSE) file for details.

qadabra/utils.py

@@ -17,7 +17,7 @@ def _validate_input(
     confounders: List[str] = None,
 ):
     logger.info("Loading metadata...")
-    md = pd.read_table(metadata, sep="\t", index_col=0)
+    md = pd.read_table(metadata, sep="\t", index_col=0, dtype={0:'str'})
 
     if factor_name not in md.columns:
         raise ValueError(f"{factor_name} not found in metadata!")
@@ -69,4 +69,4 @@ def _validate_input(
             raise ValueError("Tree tips are not a subset of table features!")
     else:
         logger.info("Reading phylogenetic tree...")
-        logger.info("(Optional tree file not provided. Skipping tree validation.)")
+        logger.info("(Optional tree file not provided. Skipping tree validation.)")

qadabra/workflow/rules/common.smk

@@ -8,7 +8,7 @@ names = datasets.index
 def get_dataset_cfg(wildcards, keys):
     d = datasets.loc[wildcards.dataset, keys].to_dict()
     if "confounders" in keys:
-        if not np.isnan(d["confounders"]):
+        if not pd.isnull(d["confounders"]):
             d["confounders"] = d["confounders"].split(";")
         else:
             d["confounders"] = []
@@ -20,7 +20,7 @@ def get_songbird_formula(wildcards):
     covariate = d["factor_name"]
     reference = d["reference_level"]
     formula = f"C({covariate}, Treatment('{reference}'))"
-    if not np.isnan(d["confounders"]):
+    if not pd.isnull(d["confounders"]):
         confounders = d["confounders"].split(";")
         formula = f"{formula} + {' + '.join(confounders)}"
     return formula
@@ -126,4 +126,4 @@ for dataset in datasets.iterrows():
             dataset=names,
             out=["empress"]
             )
-        all_input = all_input + empress_output
+        all_input = all_input + empress_output