Bioinformatics and Computational Tools for Next-Generation Sequencing Analysis in Clinical Genetics

Abstract

Clinical genetics has an important role in the healthcare system to provide a definitive diagnosis for many rare syndromes. It also can have an influence over genetics prevention, disease prognosis and assisting the selection of the best options of care/treatment for patients. Next-generation sequencing (NGS) has transformed clinical genetics making possible to analyze hundreds of genes at an unprecedented speed and at a lower price when comparing to conventional Sanger sequencing. Despite the growing literature concerning NGS in a clinical setting, this review aims to fill the gap that exists among (bio)informaticians, molecular geneticists and clinicians, by presenting a general overview of the NGS technology and workflow. First, we will review the current NGS platforms, focusing on the two main platforms Illumina and Ion Torrent, and discussing the major strong points and weaknesses intrinsic to each platform. Next, the NGS analytical bioinformatic pipelines are dissected, giving some emphasis to the algorithms commonly used to generate process data and to analyze sequence variants. Finally, the main challenges around NGS bioinformatics are placed in perspective for future developments. Even with the huge achievements made in NGS technology and bioinformatics, further improvements in bioinformatic algorithms are still required to deal with complex and genetically heterogeneous disorders.

Keywords: NGS pipeline; NGS platforms; bioinformatics; clinical genetics; high throughput data.

Figure 1

DNA sequencing timeline. Some of the most revolutionary and remarkable events in DNA sequencing. NG—next generation; PCR—polymerase chain reaction; SMS—single molecule sequencing; SeqLL—sequence the lower limit.

Figure 2

An overview of the next generation sequencing (NGS) bioinformatics workflow. The NGS bioinformatics is subdivided in the primary (blue), secondary (orange) and tertiary (green) analysis. The primary data analysis consists of the detection and analysis of raw data. Then, on the secondary analysis, the reads are aligned against the reference human genome (or de novo assembled) and the calling is performed. The last step is the tertiary analysis, which includes the variant annotation, variant filtering, prioritization, data visualization and reporting. CNV—copy number variation; ROH—runs of homozygosity, VCF—variant calling format.

Figure 3

Schematic representation of the primary analysis workflow in Ion Torrent. Briefly, the signal emitted from nucleotide incorporation is inspected by the sensor, which converts the raw voltage data into a DAT file. This file serves as input to the server, which converts into a WELLS file. This last file is used as input on the Ion Torrent Basecaller module that gives a final BAM file, ready for the secondary analysis.

Figure 4

Summary of some widely used base callers’ software available for the Illumina platform. The software is grouped according to the input file: INT (intermediate executable code) text format for the older tools and CIF (cluster intensity files) for the most recent platforms.

Figure 5

Schematic representation of the main steps involved in the post-alignment process.

Figure 6

Summary of the main methods for calling structural variants (SV) and copy number variation (CNV) from next generation sequencing (NGS) data.

Figure 7

BAM (binary alignment map) file visual inspection. Two examples of situations that may be observed through this inspection. (A) Demonstrates a case of a true-positive INDEL, confirmed by Sanger sequencing. In contrast, (B) shows a clear example of a false-positive result, where the variant is present in only reverse reads, as later demonstrated by Sanger sequencing it is a technical artifact and should be excluded from further analysis.