David L Gibbs Gibbsdavidl

## isb-cgc-query_of_the_month-november-2018.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                Gibbsdavidl
                / isb-cgc-query_of_the_month-november-2018.ipynb
            
            
              Last active
              December 6, 2018 00:05
            
              
                ISB-CGC-Query_of_the_Month-November-2018.ipynb
              
          
      Loading

      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## launch_jupyter_interface.sh
#!/bin/bash

# Made a tiny change to allow me to run it on Mac OS 10.11.6 #

# This script sets SSH tunnel and opens Chrome Browser with SOCKS proxy
# configured for direct access to the Jupyter cluster.

set -euo pipefail

DIR="${BASH_SOURCE%/*}"

## gist:5580a4cb7834e2658613bb7bec07ae49
WITH
  --
  --
  --
  geneList AS (
  SELECT
    a.entrez AS Entrez,
    Symbol AS HGNC_gene_symbol
  FROM
    `isb-cgc.QotM.Reactome_a1` a

## sift.cpp
/*M///////////////////////////////////////////////////////////////////////////////////////
//
//  IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
//  By downloading, copying, installing or using the software you agree to this license.
//  If you do not agree to this license, do not download, install,
//  copy or use the software.
//
//
//                          License Agreement

## gist:8a20097aaf8bece8fc586310795b54da
# This shows the contents of my global.R file, that contains the BigQuery SQL, googleAuthR code, and plots.


# global.R

#http://www.cookbook-r.com/Graphs/Multiple_graphs_on_one_page_(ggplot2)/
multiplot <- function(..., plotlist=NULL, file, cols=1, layout=NULL) {
  library(grid)

  # Make a list from the ... arguments and plotlist

## gist:72941f4f58c740d603c266f04939dadd
CREATE TEMP FUNCTION
  bq_linear(x ARRAY<FLOAT64>, y ARRAY<FLOAT64>)
  RETURNS FLOAT64
  LANGUAGE js AS """
    // http://tom-alexander.github.io/regression-js/

    var data = [];
    // Our data list is list of lists. The small list being each (x,y) point
    for (var i = 0; i < x.length; i++) {
      data.push([x[i],y[i]])

## gist:2de9e15afe2005a8bde393471c17db8c
SELECT
 project_short_name,
  STDDEV_SAMP( protein_expression ) / AVG( protein_expression ) AS CoV
FROM
  [isb-cgc:TCGA_hg19_data_v0.Protein_Expression]
group by
 project_short_name
order by
 CoV DESC

## gist:d5956363e9289418caf065b551b94094
WITH
  mutatedSamples AS (
  SELECT
    project_short_name,
    sample_barcode_tumor
  FROM
    `isb-cgc.TCGA_hg19_data_v0.Somatic_Mutation_MC3`
  WHERE
    Hugo_Symbol = 'EGFR'
    AND Variant_Type = 'SNP'

## gist:f72b66a3ede779c7cf3672e721baaeae
WITH
  --
  -- we start by translating the correlations that we got to ranks,
  -- based on sorting the genes on corrByGene "DESC"
  -- this will result in the highest positive correlation getting
  -- rank #1, etc
  -- we also lightly filter the genes by excluding any with near-zero
  -- or negative correlation coefficients, and the result is a list
  -- of approx 9000 genes with symbol, correlation, and rank
  geneScoresT AS (
	#!/bin/bash

	# Made a tiny change to allow me to run it on Mac OS 10.11.6 #

	# This script sets SSH tunnel and opens Chrome Browser with SOCKS proxy
	# configured for direct access to the Jupyter cluster.

	set -euo pipefail

	DIR="${BASH_SOURCE%/*}"
	WITH
	--
	--
	--
	geneList AS (
	SELECT
	a.entrez AS Entrez,
	Symbol AS HGNC_gene_symbol
	FROM
	`isb-cgc.QotM.Reactome_a1` a
	/*M///////////////////////////////////////////////////////////////////////////////////////
	//
	// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
	//
	// By downloading, copying, installing or using the software you agree to this license.
	// If you do not agree to this license, do not download, install,
	// copy or use the software.
	//
	//
	// License Agreement
	# This shows the contents of my global.R file, that contains the BigQuery SQL, googleAuthR code, and plots.


	# global.R

	#http://www.cookbook-r.com/Graphs/Multiple_graphs_on_one_page_(ggplot2)/
	multiplot <- function(..., plotlist=NULL, file, cols=1, layout=NULL) {
	library(grid)

	# Make a list from the ... arguments and plotlist
	CREATE TEMP FUNCTION
	bq_linear(x ARRAY<FLOAT64>, y ARRAY<FLOAT64>)
	RETURNS FLOAT64
	LANGUAGE js AS """
	// http://tom-alexander.github.io/regression-js/

	var data = [];
	// Our data list is list of lists. The small list being each (x,y) point
	for (var i = 0; i < x.length; i++) {
	data.push([x[i],y[i]])
	SELECT
	project_short_name,
	STDDEV_SAMP( protein_expression ) / AVG( protein_expression ) AS CoV
	FROM
	[isb-cgc:TCGA_hg19_data_v0.Protein_Expression]
	group by
	project_short_name
	order by
	CoV DESC
	WITH
	mutatedSamples AS (
	SELECT
	project_short_name,
	sample_barcode_tumor
	FROM
	`isb-cgc.TCGA_hg19_data_v0.Somatic_Mutation_MC3`
	WHERE
	Hugo_Symbol = 'EGFR'
	AND Variant_Type = 'SNP'
	WITH
	--
	-- we start by translating the correlations that we got to ranks,
	-- based on sorting the genes on corrByGene "DESC"
	-- this will result in the highest positive correlation getting
	-- rank #1, etc
	-- we also lightly filter the genes by excluding any with near-zero
	-- or negative correlation coefficients, and the result is a list
	-- of approx 9000 genes with symbol, correlation, and rank
	geneScoresT AS (