microRNA Gene Expression Prediction Model Pipeline

Introduction

• A pipeline is used to build microRNA gene expression prediction models

Requirements

1. Python2.7
2. R 3.0+
3. data.table
4. glmnet
5. qvalue
6. dplyr
7. bit64
8. doMC

Installing Pipeline

git clone https://github.com/jiamaozheng/microRNA_gene_expression_prediction_model_pipeline

Supported Input File Formats

• expression file - miRNA_CTR_Exp.RDS
• expression annotation file - miRBase_miRNA_gene_annotation.RDS
• genotype file - geuvadis_snps.txt
• SNP annotation file - gEUVADIS.SNP.annotation.RDS

Command Line Parameters

Argument	Abbre	Required	Default	Description
--project_name	-p	Yes	None	Project name (e.g. gEUVADIS_LCL, TCGA or Framingham)
--molecular_type	-m	Yes	None	Molecular types (e.g. mRNA, miRNA or shRNA)
--alpha	-a	Yes	None	Alpha values (e.g. 0.05, 0.5, or 1)
--snpset	-s	Yes	None	SNP set used for analysis (e.g. 1KG_snps, HapMap_snps)
--n_k_folds	-n	No	10	The number of folds for cross-validation (e.g. 10)
--fdr_level	-f	No	0.05	FDR used for filtering modelS (e.g. 0.05)
--window	-w	No	1e6	The number of bps to +/- transcription start site(TSS)
--expression_path	-e	No	''	User-defined expression file path
--genotype_path	-g	No	''	User-defined genotype file path
--gene_annot_path	-x	No	''	User-defined gene annotation file path
--snp_annot_path	-y	No	''	User-defined snp annotation file path
--intermediate_path	-i	No	''	User-defined intermediate file path
--results_output_path	-r	No	''	User-defined output file path

Running Pipeline

• Example 1 (basic command, recommended): Navigate to the folder that contains downloaded pipeline microRNA_gene_expression_prediction_model_pipeline, create a default input directory by using a script, navigate to the pipeline folder microRNA_gene_expression_prediction_model_pipeline, and then execute the following command with four requried parameters.

	./run.py \
	--project_name gEUVADIS_LCL_miRNA \
	--molecular_type miRNA \
	--alpha 0.5 \
	--snpset 1KG_snps \

Alternatively, you can fun the following shortcut

	./run.py -p gEUVADIS_LCL_miRNA -m miRNA -a 0.5 -s 1KG_snps 

• Example 2 (basic command + usered-defined input file paths): Navigate to the pipeline folder microRNA_gene_expression_prediction_model_pipeline, and then execute the following command with four requried parameters and five user-defined input file option parameters

	./run.py \
	--project_name gEUVADIS_LCL_miRNA \
	--molecular_type miRNA \
	--alpha 0.5 \
	--snpset 1KG_snps \

	--expression_path ../input_mirna/expression_phenotypes/miRNA_CTR_Exp.RDS \
	--genotype_path ../input_mirna/genotype/ \
	--gene_annot_path ../input_mirna/gene_annotation/miRBase_miRNA_gene_annotation.RDS \
	--snp_annot_path ../input_mirna/snp_annotation/gEUVADIS.SNP.annotation.RDS \
	--intermediate_path ../input_mirna/intermediate/

• Example 3 (basic command + user-defined input file paths + user-provided output file path): Navigate to the pipeline folder microRNA_gene_expression_prediction_model_pipeline, and then execute the following command with four requried parameters, five user-defined input file option parameters, and one user-defined output file option parameter

	./run.py \
	--project_name gEUVADIS_LCL_miRNA \
	--molecular_type miRNA \
	--alpha 0.5 \
	--snpset 1KG_snps \

	--expression_path ../input_mirna/expression_phenotypes/miRNA_CTR_Exp.RDS \
	--genotype_path ../input_mirna/genotype/ \
	--gene_annot_path ../input_mirna/gene_annotation/miRBase_miRNA_gene_annotation.RDS \
	--snp_annot_path ../input_mirna/snp_annotation/gEUVADIS.SNP.annotation.RDS \
	--intermediate_path ../input_mirna/intermediate/ \ 
	--results_output_path ../output_mirna/

• Example 4 (basic command + user-defined input file paths + user-defined output file path + default parameters): Navigate to the pipeline folder microRNA_gene_expression_prediction_model_pipeline, and then execute the following command with four requried parameters, five user-defined input file option parameters, one user-defined output file option parameter, and three default parameters which can be modified by users if necessary

	./run.py \
	--project_name gEUVADIS_LCL_miRNA \
	--molecular_type miRNA \
	--alpha 0.5 \
	--snpset 1KG_snps \

	--expression_path ../input_mirna/expression_phenotypes/miRNA_CTR_Exp.RDS \
	--genotype_path ../input_mirna/genotype/ \
	--gene_annot_path ../input_mirna/gene_annotation/miRBase_miRNA_gene_annotation.RDS \
	--snp_annot_path ../input_mirna/snp_annotation/gEUVADIS.SNP.annotation.RDS 
	--intermediate_path ../input_mirna/intermediate/

	--n_k_folds 10 
	--fdr_level 0.05 
	--window 1e6