look up genes by gene_id

app.py

@@ -11,6 +11,7 @@ st.markdown("""
 **CryptoCEN** is a co-expression network for *Cryptococcus neoformans* built on 1,524 RNA-seq runs across 34 studies.
 A pair of genes are said to be co-expressed when their expression is correlated across different conditions and
 is often a marker for genes to be involved in similar processes. 
+
 To Cite:
 MJ O'Meara, JR Rapala, CB Nichols, C Alexandre, B Billmyre, JL Steenwyk, A Alspaugh,
 TR O'Meara CryptoCEN: A Co-Expression Network for Cryptococcus neoformans reveals
@@ -22,6 +23,11 @@ novel proteins involved in DNA damage repair
 Put in the ``CNAG_#####`` gene_id for two genes.
 """)
 
+h99_transcript_annotations = datasets.load_dataset(
+    path = "maomlab/CryptoCEN",
+    data_files = {"h99_transcript_annotations": "h99_transcript_annotations.tsv"})
+h99_transcript_annotations = h99_transcript_annotations["h99_transcript_annotations"].to_pandas()
+
 estimated_expression_meta = datasets.load_dataset(
     path = "maomlab/CryptoCEN",
     data_files = {"estimated_expression_meta": "Data/estimated_expression_meta.tsv"})
@@ -32,6 +38,12 @@ estimated_expression = datasets.load_dataset(
     data_files = {"estimated_expression": "estimated_expression.tsv"})
 estimated_expression = estimated_expression["estimated_expression"].to_pandas()
 
+print(f"estimated_expression shape: {estimated_expression.shape}")
+
+print(f"transcript_annotations are equal: {sum(h99_transcript_annotations['cnag_id'] == estimated_expression.index)}")
+
+
+
 col1, col2, col3 = st.columns(spec = [0.2, 0.2, 0.6])
 with col1:
     gene_id_1 = st.text_input(
@@ -40,6 +52,7 @@ with col1:
         max_chars = 10,
         help = "CNAG Gene ID e.g. CNAG_04365")
 
+
 with col2:
     gene_id_2 = st.text_input(
         label = "Gene ID 2",
@@ -48,8 +61,8 @@ with col2:
         help = "CNAG Gene ID e.g. CNAG_04222")
 
 chart_data = pd.DataFrame({
-    "expression_1": np.log10(estimated_expression.loc[estimated_expression.index == gene_id_1].to_numpy()[0] + 1),
-    "expression_2": np.log10(estimated_expression.loc[estimated_expression.index == gene_id_2].to_numpy()[0] + 1),
+    "expression_1": np.log10(estimated_expression.loc[h99_transcript_annotations["gene_id"] == gene_id_1].to_numpy()[0] + 1),
+    "expression_2": np.log10(estimated_expression.loc[h99_transcript_annotations["gene_id"] == gene_id_2].to_numpy()[0] + 1),
     "run_accession": estimated_expression.columns,
     "run_accession_meta": estimated_expression_meta["run_accession"],
     "study_accession": estimated_expression_meta["study_accession"]})