Next generation sequencing technology and genomewide data analysis: Perspectives for retinal research

Abstract

The advent of high throughput next generation sequencing (NGS) has accelerated the pace of discovery of disease-associated genetic variants and genomewide profiling of expressed sequences and epigenetic marks, thereby permitting systems-based analyses of ocular development and disease. Rapid evolution of NGS and associated methodologies presents significant challenges in acquisition, management, and analysis of large data sets and for extracting biologically or clinically relevant information. Here we illustrate the basic design of commonly used NGS-based methods, specifically whole exome sequencing, transcriptome, and epigenome profiling, and provide recommendations for data analyses. We briefly discuss systems biology approaches for integrating multiple data sets to elucidate gene regulatory or disease networks. While we provide examples from the retina, the NGS guidelines reviewed here are applicable to other tissues/cell types as well.

Keywords: ChIP-seq; Chromatin; Epigenetics; Gene regulatory network; Genomics; High throughput data; NGS data integration; Network analysis; Photoreceptor; RNA-Seq; Retina; Retinal disease; Systems biology; Transcriptome; Vision; Whole exome sequencing; Whole genome sequencing.

Figure 1. Timeline of human genetics and genomic technologies

NGS based applications have been utilized widely in vision and other biomedical research. A. From the discovery of DNA molecule until today, substantial scientific and technical advancements in human genetics and eye field are presented in a chronological order. The first NGS report was published a decade after the launch of human genome project. B. Cumulative number of biomedical research papers based on NGS technologies from 2008 to 2015 in PubMed database. We believe the number of scientific reports based on NGS technologies will continue to increase as NGS become more available and affordable. C. Profiling genomic variations is more employed than expression and genome binding profiling in vision research studies. As of December 2015, the total number of NGS based studies doubled in the eye field compared to two years ago. PCR, polymerase chain reaction; RP, retinitis pigmentosa; HGP, human genome project; NGS, next generation sequencing; AMD, age-related macular degeneration; GWAS, genomewide association study.

Figure 2. Illumina sequencing and data processing workflow

A. Denaturated NGS library fragments are flowed across a flow cell and hybridize on a lawn of complementary Illumina adapter oligos. Complementary fragments are extended, amplified via bridge amplification PCR, and denaturated, resulting in clusters of identical single-stranded library fragments. B. Fragments are primed and sequenced utilizing reversible terminator nucleotides. Base pairs are identified after laser excitation and fluorescence detection. C. Raw data is demultiplexed into individual libraries and assessed for quality. Removing adapter reads reduces technical noise. Finally reads are aligned onto assembly of interest.

Figure 3. WES workflow and analysis

Genomic DNA from cells or tissue is tagmented using hyperactive Tn5 transposase coupled with Illumina sequencing adapters as described in (http://www.illumina.com/products/nextera-rapid-capture-custom-enrichment-kit.html). After PCR amplification, DNA probes specific to exonic sequences are used to isolate coding sequences using two-step hybridization. Library amplification with index primers allow for multiplexing a variety of libraries in the same sequencing flow cell. After sequencing and read mapping steps, PCR duplicates are removed using available computational tools. Realignment around indels, base recalibration, variant calling and annotations are all WES-specific computational processes to extract variant information.

Figure 4. RNA sequencing workflow and analysis

Total RNA is extracted and ribosomal RNA is either removed to enrich for other RNA species, or polyA-tailed RNA are isolated using poly(T) oligomer magnetic beads as described in (http://www.illumina.com/products/truseq_stranded_total_rna_library_prep_kit.html). RNA is then fragmented using sonication, followed by cDNA synthesis, end repair, adapter ligation, and indexing. After PCR amplification and library quantification, RNA reads are mapped to known transcripts and the whole genome to facilitate transcript identification and quantification. Multiple secondary analyses exist to understand the expression profile of cells and whole tissues.

Figure 5

Construction of co-expression networks and functional enrichment of network modules A. Co-expressed genes (also called linked genes) can be identified and grouped into network modules (represented with different colors). Here we demonstrate example network structures of three network modules (red, blue and turquoise), built using Weighted Gene Co-expression Network Analysis tool (WGCNA)(Langfelder and Horvath, 2008). In each network module, genes are represented as nodes and co-expressed genes are linked. In co-expression networks, it is believed that highly connected genes (also called hub genes) represent biologically significant genes since their dysregulation may affect many other linked genes. B. Genes with similar expression patterns tend to group in the same network module, and are more likely to be in the similar biological processes/pathways. Biological relevance of network modules can be elucidated with diverse online functional enrichment analysis tools. Here we show the first two most significant Gene Ontology (GO) terms related to each module after functional GO enrichment analysis using DAVID online tool (Huang da et al., 2009). Additionally, heatmap shows expression patterns of genes in each network module across time points.

Figure 6. ChIP and ATAC sequencing workflow and analysis

Retina is dissected from the eye, and cells are dissociated for follow-up studies. A. (ChIP-seq): Isolated chromatin is fixed (or unfixed) and fragmented as described in (Sheaffer and Schug, 2016). Magnetic beads conjugated to antibody specific to the target protein are used to precipitate fragments bound to the protein of interest. Adapter ligation and indexing is followed by sequencing and quality control assessment. Reads are mapped onto the genome assembly and PCR duplicates are removed to minimize experimental artifacts. Mitochondrial DNA may also be removed to improve data set comparisons. Peak calling using one of the many available tools and DNA-binding sequencing motif analysis may facilitate the elucidation of the genome binding profile for the precipitated protein. B. (ATAC-seq): Unfixed chromatin is obtained and tagmented as described in (Buenrostro et al., 2013). PCR amplification with index primers facilitates multiplexing strategies. Sequencing and mapping are followed by PCR duplicate and mitochondrial DNA removal using available bioinformatics tools or custom scripts. Open chromatin peak calling allows for the identification of active regulatory regions and may be combined with other NGS data sets to improve systems integration analyses.

Figure 7. Integrating NGS data for systems-level understanding

Knowledge integration resulting from analyzing sequencing data from genomics, epigenomics, transcriptomics, and literature can help decode networks and variants underlying development and disease. A systems level study can yield better identification of disease genes and drug targets.