From 86a687788c28b938d5baccc5f64428683f1c3359 Mon Sep 17 00:00:00 2001
From: Dylan Hoogduin <dylanhoogduin@Dylans-MacBook-Pro.local>
Date: Fri, 17 Apr 2020 15:35:21 +0200
Subject: [PATCH] add material views

---
 MaterialViews.py         | 39 +++++++++++++++++++++++++++++++++++++++
 Seeder.py                |  9 ++++++++-
 mviews/transcript_mv.sql |  7 +++++++
 3 files changed, 54 insertions(+), 1 deletion(-)
 create mode 100644 MaterialViews.py
 create mode 100644 mviews/transcript_mv.sql

diff --git a/MaterialViews.py b/MaterialViews.py
new file mode 100644
index 0000000..548b652
--- /dev/null
+++ b/MaterialViews.py
@@ -0,0 +1,39 @@
+import sqlalchemy
+
+
+class MaterialViews():
+
+    def __init__(self, db):
+        self.db = db
+        self.run_mv_transcript()
+
+    def run_mv_transcript(self):
+        with open('mviews/transcript_mv.sql') as file:
+            query = file.read()
+        self.prepare(query)
+        result = self.db.execute("SELECT id FROM tissue")
+        tissues = [item for item in result]
+        for tissue in tissues:
+            tissue_id = tissue[0]
+            query = "SELECT  gene, tissue, avg(count) from transcript "
+            query += "WHERE tissue = {tissue} ".format(tissue=tissue_id)
+            query += "GROUP BY gene "
+            query += "ORDER BY avg(count) DESC "
+            query += "LIMIT 100 "
+            result = self.db.execute(query)
+            for item in result:
+                query = "INSERT INTO transcript_mv (gene, tissue, count_avg) "
+                query += "VALUES ({x[0]}, {x[1]}, {x[2]})".format(x=list(item))
+                self.db.execute(query)
+        self.db.close()
+
+    def prepare(self, file):
+        sql_command = ''
+        for line in file:
+            if not line.startswith('--') and line.strip('\n'):
+                sql_command += line.strip('\n')
+                if sql_command.endswith(';'):
+                    try:
+                        self.db.execute(sqlalchemy.text(sql_command))
+                    finally:
+                        sql_command = ''
\ No newline at end of file
diff --git a/Seeder.py b/Seeder.py
index 6c3fc5a..044eb9f 100644
--- a/Seeder.py
+++ b/Seeder.py
@@ -1,5 +1,6 @@
 import os, time, sys
 from Database import db
+from MaterialViews import MaterialViews
 from Merger import GeneAliasRetriever
 from sqlalchemy.exc import SQLAlchemyError
 from dotenv import load_dotenv
@@ -150,6 +151,9 @@ def insert_count(self, stage_file, tissue_file):
         print('--- All counts inserted! ---')
         self.connection.close()
 
+    def run_mv(self):
+        self.connect_to_db(self.database)
+        MaterialViews(self.connection)
 
 if __name__ == '__main__':
     DB_DIALECT = os.getenv("DB_DIALECT")
@@ -163,6 +167,7 @@ def insert_count(self, stage_file, tissue_file):
     INSERT_REF = os.getenv("INSERT_REF")
     INSERT_COUNT = os.getenv("INSERT_COUNT")
     CORRECT_GENES = os.getenv("CORRECT_GENES")
+    RUN_MV = os.getenv("RUN_MV")
     seeder = Seeder(
         dialect=DB_DIALECT,
         driver=DB_DRIVER,
@@ -178,4 +183,6 @@ def insert_count(self, stage_file, tissue_file):
     if INSERT_COUNT == "true":
         seeder.insert_count(stage_file='datasets/stage.csv', tissue_file='datasets/tissue.csv')
     if CORRECT_GENES == "true":
-        seeder.correct_genes("updated_genes.txt")
\ No newline at end of file
+        seeder.correct_genes("updated_genes.txt")
+    if RUN_MV == "true":
+        seeder.run_mv()
\ No newline at end of file
diff --git a/mviews/transcript_mv.sql b/mviews/transcript_mv.sql
new file mode 100644
index 0000000..ad18a2b
--- /dev/null
+++ b/mviews/transcript_mv.sql
@@ -0,0 +1,7 @@
+DROP TABLE IF EXISTS transcript_mv;
+CREATE TABLE transcript_mv (
+    gene INT NOT NULL
+  , tissue INT NOT NULL
+  , count_avg    DECIMAL(10,2) NOT NULL
+  , UNIQUE INDEX product (gene, tissue)
+);
\ No newline at end of file