QTL analysis

QTL stands for Quantitative Trait Locus (or Loci). QTL Analysis is to measure association of a genetic variant (SNV or SNP) with a particular quantitative trait, e.g. phenotype (height, blood pressure), or molecular phenotype (gene expression, protein expression) etc. The input data should be genotype matrix data node and matrix of quantitative traits on the same set of subjects.

Click on a data node contains genotype matrix to choose Statistics > QTL analysis on the menu

It will open a configuration dialog:

Click on Select data node button to select a node containing quantitative trait matrix, e.g. normalized gene expression data node, normalized protein expression data node etc.

After the protein data node, click Next in the dialog

The task is based on ANOVA¹ to test association of every variant genotype with every phenotype feature. Genotype information is treated as numeric, in other words, the method is to test additive effect of the variant on the abundance of the phenotype feature. User can also include any other sample level attributes as covariates in the model.

Click Next to configure the output threshold:

Click Finish to run the task.

Double click on the QTL analysis output data node to open the report

In the report table, each row is a variant and phenotype feature pair, any additional fields from the phenotype annotation can be retrieved from the Optional columns on the upper-right corner of the table.

In the QTL results section:

• pvalue: statistical significance of the association, smaller pvalue means more significant

• FDR: false discovery rate based on Benjamini-Hochberg procedure²

• statistics: t-statistic

• beta: effect size estimate indicate how much expression changes per allele change. Positive value means alternate allele increases expression, negative value means alternate allele decreases expression, magnitude indicates strength of effect.

Reference

1. Shabalin, A.A. Matrix eQTL: Ultra fast eQTL analysis via large matrix operations. Bioinformatics 28, no. 10 (2012): 1353-1358. https://www.rdocumentation.org/packages/MatrixEQTL/versions/1.2.0/topics/Matrix_eQTL_engine

2. https://www.jstor.org/stable/2346101