457
spanning entire mRNA transcripts, discerning gene isoforms (Goodwin et al. 2016,
Tyson et al. 2018).
The ideal situation for genomic research normally requires a reference genome
and re-sequencing or genotyping data for hundreds to thousands of markers. For
non-model species, however, reference genome and re-sequencing are more difficult to achieve due to their high investment or methodological issues (e.g. large and/
or repetitive genome). The advances on these HTS technologies have also driven
down the costs of DNA sequencing allowing the development of massive genotyping methods. These methods are particularly useful for species without any genomic
data, for which previously the development and genotyping of molecular markers
[e.g. simple sequence repeats (SSRs), single-nucleotide polymorphisms (SNPs)]
was traditionally performed by using relatively expensive approaches (e.g. Eckert
et al. 2013; Geraldes et al. 2013; Pavy et al. 2013).
One of the massive genotyping methods that emerged in recent years is called
reduced representation, as it is focused only on a portion of the individuals’ genome
which reduces sequencing costs. Particularly, restriction enzyme-guided sequencing
approaches have become the most popular reduced representation methods for nonmodel organisms (Parchman et al. 2018), being restriction site-associated DNA
sequencing (RADseq, Baird et al. 2008; ddRADseq, Peterson et al. 2012), and genotyping by sequencing (GBS, Elshire et al. 2011) the most popular strategies. Briefly,
in these methods, library preparation consists of genomic DNA digestion with one or
two restriction enzymes, subsequent adapter ligation, selection, and high-throughput
sequencing of DNA fragments. This approach simultaneously allows genotyping a
large number of molecular markers (in particular SNPs) in a large number of individuals in the same assay. Massive genotyping methods prompted, therefore,
genomic research in non-model species as they can be easily applied to any species
for rapid marker discovery and genotyping without requiring previous genomic
resources (Davey et al. 2011; Andrews et al. 2016; Parchman et al. 2018).
With the explosion of these enormous new datasets, the need for better and faster
analyses increased dramatically. Bioinformatics is a discipline with a great expansion in the last decade as a way to deal with all these new approaches in genomic,
transcriptomic and other -omics. The main objective in this new era is to use these
large datasets, accounting for their biases and sequencing errors, to extract useful
information and to gain better knowledge of the studied subject. In summary, to go
from data to information is not an easy task to fulfill and is one of the major bottlenecks in -omics research (Greene et al. 2014).
The bioinformatics area is interdisciplinary from its beginning, going from database management, sequence analysis and/or statistics in genomics. In this new era,
researchers can establish new experimental designs and tackle never-before-tried
biological challenges, such as how do all genes change expression under a new
condition, or genotype millions of positions in a genome for which no reference
genome exist (Feuillet et al. 2011). These new technologies have not only scaled up
datasets but have radically “modified/changed” the traditional paradigm of hypothesis-driven science, with a new paradigm of big data science (Stephens et al. 2015).
Data science is a specialized field that combines multiple areas such as statistics,
mathematics, intelligent data capture techniques, data cleansing, mining, and
17 Application of High-Throughput Sequencing Technologies in Native Forest Tree…
spanning entire mRNA transcripts, discerning gene isoforms (Goodwin et al. 2016,
Tyson et al. 2018).
The ideal situation for genomic research normally requires a reference genome
and re-sequencing or genotyping data for hundreds to thousands of markers. For
non-model species, however, reference genome and re-sequencing are more difficult to achieve due to their high investment or methodological issues (e.g. large and/
or repetitive genome). The advances on these HTS technologies have also driven
down the costs of DNA sequencing allowing the development of massive genotyping methods. These methods are particularly useful for species without any genomic
data, for which previously the development and genotyping of molecular markers
[e.g. simple sequence repeats (SSRs), single-nucleotide polymorphisms (SNPs)]
was traditionally performed by using relatively expensive approaches (e.g. Eckert
et al. 2013; Geraldes et al. 2013; Pavy et al. 2013).
One of the massive genotyping methods that emerged in recent years is called
reduced representation, as it is focused only on a portion of the individuals’ genome
which reduces sequencing costs. Particularly, restriction enzyme-guided sequencing
approaches have become the most popular reduced representation methods for nonmodel organisms (Parchman et al. 2018), being restriction site-associated DNA
sequencing (RADseq, Baird et al. 2008; ddRADseq, Peterson et al. 2012), and genotyping by sequencing (GBS, Elshire et al. 2011) the most popular strategies. Briefly,
in these methods, library preparation consists of genomic DNA digestion with one or
two restriction enzymes, subsequent adapter ligation, selection, and high-throughput
sequencing of DNA fragments. This approach simultaneously allows genotyping a
large number of molecular markers (in particular SNPs) in a large number of individuals in the same assay. Massive genotyping methods prompted, therefore,
genomic research in non-model species as they can be easily applied to any species
for rapid marker discovery and genotyping without requiring previous genomic
resources (Davey et al. 2011; Andrews et al. 2016; Parchman et al. 2018).
With the explosion of these enormous new datasets, the need for better and faster
analyses increased dramatically. Bioinformatics is a discipline with a great expansion in the last decade as a way to deal with all these new approaches in genomic,
transcriptomic and other -omics. The main objective in this new era is to use these
large datasets, accounting for their biases and sequencing errors, to extract useful
information and to gain better knowledge of the studied subject. In summary, to go
from data to information is not an easy task to fulfill and is one of the major bottlenecks in -omics research (Greene et al. 2014).
The bioinformatics area is interdisciplinary from its beginning, going from database management, sequence analysis and/or statistics in genomics. In this new era,
researchers can establish new experimental designs and tackle never-before-tried
biological challenges, such as how do all genes change expression under a new
condition, or genotype millions of positions in a genome for which no reference
genome exist (Feuillet et al. 2011). These new technologies have not only scaled up
datasets but have radically “modified/changed” the traditional paradigm of hypothesis-driven science, with a new paradigm of big data science (Stephens et al. 2015).
Data science is a specialized field that combines multiple areas such as statistics,
mathematics, intelligent data capture techniques, data cleansing, mining, and
17 Application of High-Throughput Sequencing Technologies in Native Forest Tree…