Introduction

In this guide, we discuss some of the challenges involved in human variant analysis and explore some of the solutions available for addressing these challenges. Whether you are working with a core facility, a bioinformatics group, or doing variant analysis on your own, this guide will help you learn about important considerations to keep in mind throughout the process.

Why study variants?

A common workflow in the study of human genetic variation involves the analysis and identification of deleterious variants or of variants associated with a particular population or trait. There are thousands of known variants that cause Mendelian disorders, and thousands more whose molecular basis is yet unknown.

In a typical human variant analysis study, the researcher’s goal is to identify which single-nucleotide polymorphisms (SNPs), small insertions and deletions (INDELs), copy number variations (CNVs), or other types of structural variations and rearrangements (SVs) have functional significance. Functionally significant variants are those that cause amino acid changes, abnormal exon splicing, or other protein structure changes that contribute to a diseased state.

Challenges in variant analysis

The path from raw DNA reads to an understanding of the clinical significance of their variants can be long and complex. Each step and application will affect the accuracy and completeness of the results. An additional complication is that the large number of tools available have led to a myriad of pipelines. Just analyzing one data set can require mastery of up to a dozen bioinformatics applications and online databases.

Another challenge is that many open-source tools are created and then abandoned as graduate students finish their bioinformatics dissertations or laboratories lose funding or change focus. Few of these tools provide comprehensive documentation or instruction.

Even supposedly “automated” open source and commercial pipelines may rely on command line utilities at some of the key steps or require you to navigate through multiple interfaces at each step in the pipeline. These additional steps generate multiple intermediate data files and increase hands-on time and clock time to get from raw sequencing data to the data analysis steps. It is therefore important to assess your ability and willingness to enter a pipeline that may require the use of command line tools and at least an intermediate knowledge of bioinformatics.