458
programming to prepare and align big data for intelligent analysis to extract insights
and information from them. Moreover, specifically in functional genomic analysis
of species lacking a reference genome (de novo), this involves the integration of
heterogeneous data sources including (but not limited to) transcriptome assembly
and annotation, along with gene expression and genetic variants. In particular, it is
important to establish that the volume and complexity of the data produced imply a
technical, but also a scientific, challenge (Greene et al. 2014). Currently, many projects use big data HTS experiments, analyse the results and generate new hypotheses
that are then further tested. Some strategies used in tree breeding which are of great
impact, later covered in this chapter, are:
1. Novel SNP discovery and genome-wide genotyping of hundreds of individuals
(De Pristo et al. 2011). With variant discovery, nowadays, researchers can
sequence DNA or RNA samples from different individuals and without previous
knowledge determine variant positions and genotype their samples with thousands of newly discovered SNPs. This opens a wide range of possibilities, never
achieved before at such low cost.
2. De novo genomic and transcriptome assembly (Birol et al. 2013; Conesa et al.
2016). In de novo genomic or transcriptome assembly, researchers can identify
new (near full length) transcripts or genes (loci) and provide functional annotations to better characterize their material, all with a lot less effort than ever
thought few decades ago.
3. Reduced representation sequencing is a technique used to sequence an abridged
representation of the genome in many individuals. Ideally, it allows sequencing
the same reduced regions of the genome in all individuals and therefore identify
variants in those regions.
4. Genome-wide association studies (GWAS) to identify associations between a
genetic variant and a specific phenotype. It is basically a statistical correlation
between a marker and a phenotype (Brachi et al. 2011; Beilsmith et al. 2019).
5. Genomic selection is a strategy which simultaneously estimates the effects of all
available markers along the genome to predict individually breeding values,
without specific knowledge of the genes that contribute to a particular character
(Sewell and Neale 2000; Neale 2007).
17.2 Applications of HTS Technologies on Forest
Tree Species
Genomics research in forest tree species has been prompted by the development of
HTS technologies and bioinformatics tools. A systematic literature search from
Scopus database combining two search word strategies
1
for the period 2016–2019
1 Strategy A: (TITLE-ABS-KEY (“next-generation sequencing” OR “massive sequencing” OR
“high-throughput sequencing”) AND TITLE-ABS-KEY (“forest tree” OR “tree species” OR
S. L. Torales et al.
programming to prepare and align big data for intelligent analysis to extract insights
and information from them. Moreover, specifically in functional genomic analysis
of species lacking a reference genome (de novo), this involves the integration of
heterogeneous data sources including (but not limited to) transcriptome assembly
and annotation, along with gene expression and genetic variants. In particular, it is
important to establish that the volume and complexity of the data produced imply a
technical, but also a scientific, challenge (Greene et al. 2014). Currently, many projects use big data HTS experiments, analyse the results and generate new hypotheses
that are then further tested. Some strategies used in tree breeding which are of great
impact, later covered in this chapter, are:
1. Novel SNP discovery and genome-wide genotyping of hundreds of individuals
(De Pristo et al. 2011). With variant discovery, nowadays, researchers can
sequence DNA or RNA samples from different individuals and without previous
knowledge determine variant positions and genotype their samples with thousands of newly discovered SNPs. This opens a wide range of possibilities, never
achieved before at such low cost.
2. De novo genomic and transcriptome assembly (Birol et al. 2013; Conesa et al.
2016). In de novo genomic or transcriptome assembly, researchers can identify
new (near full length) transcripts or genes (loci) and provide functional annotations to better characterize their material, all with a lot less effort than ever
thought few decades ago.
3. Reduced representation sequencing is a technique used to sequence an abridged
representation of the genome in many individuals. Ideally, it allows sequencing
the same reduced regions of the genome in all individuals and therefore identify
variants in those regions.
4. Genome-wide association studies (GWAS) to identify associations between a
genetic variant and a specific phenotype. It is basically a statistical correlation
between a marker and a phenotype (Brachi et al. 2011; Beilsmith et al. 2019).
5. Genomic selection is a strategy which simultaneously estimates the effects of all
available markers along the genome to predict individually breeding values,
without specific knowledge of the genes that contribute to a particular character
(Sewell and Neale 2000; Neale 2007).
17.2 Applications of HTS Technologies on Forest
Tree Species
Genomics research in forest tree species has been prompted by the development of
HTS technologies and bioinformatics tools. A systematic literature search from
Scopus database combining two search word strategies
1
for the period 2016–2019
1 Strategy A: (TITLE-ABS-KEY (“next-generation sequencing” OR “massive sequencing” OR
“high-throughput sequencing”) AND TITLE-ABS-KEY (“forest tree” OR “tree species” OR
S. L. Torales et al.