From 1f27d5e0f6c65908d021d0c38f0b2e0671a63f5d Mon Sep 17 00:00:00 2001
From: Daniel Himmelstein <daniel.himmelstein@gmail.com>
Date: Wed, 5 Oct 2016 18:25:03 -0400
Subject: [PATCH] Rerun with fixed entrez_gene_id bug

---
 6.differential-expression.ipynb      | 12 ++++++------
 scripts/6.differential-expression.py | 18 +++++++++++++-----
 2 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/6.differential-expression.ipynb b/6.differential-expression.ipynb
index b0355c7..deca7d5 100644
--- a/6.differential-expression.ipynb
+++ b/6.differential-expression.ipynb
@@ -192,7 +192,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 5,
    "metadata": {
     "collapsed": false
    },
@@ -233,7 +233,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 6,
    "metadata": {
     "collapsed": false
    },
@@ -244,7 +244,7 @@
        "22973"
       ]
      },
-     "execution_count": 39,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -259,7 +259,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 7,
    "metadata": {
     "collapsed": false
    },
@@ -364,7 +364,7 @@
        "349009 -1.196275        0.557971   SLC25A5-AS1  "
       ]
      },
-     "execution_count": 42,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -376,7 +376,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 8,
    "metadata": {
     "collapsed": false
    },
diff --git a/scripts/6.differential-expression.py b/scripts/6.differential-expression.py
index 655969f..c0a02c0 100644
--- a/scripts/6.differential-expression.py
+++ b/scripts/6.differential-expression.py
@@ -57,7 +57,7 @@ def get_diffex(subtype_df):
     ttest = ttest_1samp(diffex_df, popmean=0, axis=0)
 
     df = pandas.DataFrame.from_items([
-        ('entrez_gene_id', diffex_df.columns.astype(int)),
+        ('entrez_gene_id', diffex_df.columns),
         ('patients', len(diffex_df)),
         ('tumor_mean', tumor_df.mean()),
         ('normal_mean', normal_df.mean()),
@@ -67,21 +67,29 @@ def get_diffex(subtype_df):
     ])
     return df
 
-diffex_df = type_df.groupby('acronym').apply(get_diffex).reset_index('acronym')
+diffex_df = (type_df
+    .groupby('acronym')
+    .apply(get_diffex)
+    .reset_index('acronym')
+    .query("patients >= 5")
+)
+
+diffex_df.entrez_gene_id = diffex_df.entrez_gene_id.astype(int)
 
 
 # In[6]:
 
 # Add gene symbols
 path = os.path.join('data', 'genes.tsv')
-gene_df = pandas.read_table(path)
+gene_df = pandas.read_table(path, low_memory=False)
 gene_df = gene_df[['entrez_gene_id', 'symbol']]
-diffex_df = gene_df.merge(diffex_df, how='right')
+len(gene_df)
 
 
 # In[7]:
 
-diffex_df.head()
+diffex_df = diffex_df.merge(gene_df, how='left')
+diffex_df.tail()
 
 
 # In[8]: