Table 1.

Next generation sequencing techniques used for studying gene expression and gene regulation in evolutionary studies. Methods labelled with * require a reference genome.

Method	Key information
RNAseq	Summary: RNA is isolated and reverse transcribed into cDNA for library preparation and sequencing. Practical considerations: The most common protocol uses oligo-dT primers to enrich for polyadenylated RNAs for reverse transcription of processed mRNA [17] and the majority of lncRNAs [118]. Alternative protocols use total RNA and ribosome depletion prior to reverse transcription with random oligos to obtain other RNA molecules (e.g., immature mRNA, miRNA, and siRNA) [119]. For small RNA enrichment several commercial kits are available to select for molecule sizes less than 30 nucleotides [120]. Applications: Transcriptome generation for gene annotation including alternative isoforms (paired-end sequencing) and differential gene expression analysis between different samples (e.g., tissues, experimental conditions, populations of the same species or even species showing different phenotypes) [18,19,20,21,121]. RNAseq is also a useful tool for miRNA profiling and annotation [122] as well as differential expression of lncRNAs [123]. Single cell application: [124,125,126]
ATACseq*	Summary: Accessible chromatin regions which are not condensed by histones, are digested with a genetically modified transposase (Tn5). Nucleotide overhangs (tagmentation) are utilized for specific adapter ligation during the library preparation and sequencing [69,127]. This method substituted previous ones such as DNaseseq and FAIREseq, due to its simplicity and effectiveness. Practical considerations: Usually the protocol should be done with fresh tissue and a defined number of nuclei/cells (e.g 500–50,000 for mammalian tissues [127]) that have to be estimated prior to tagmentation. These technical aspects limit the number of samples that can be processed simultaneously. However, protocols were successfully applied to frozen tissue [128]. Applications: ATACseq is commonly used to complement RNAseq data to identify potential regulatory regions (enhancers) [129]. ATACseq can also be used to evaluate chromatin structure dynamics and epigenetic changes by providing information about histone position as well as a complementary approach to ChIPseq to characterize transcription factor and repressor (e.g., CTCF) occupancies [69]. Single cell application: [130,131]
ChIPseq*	Summary: DNA bound proteins (e.g., transcription factors, histones) are crosslinked and the chromatin is digested with restriction enzymes. Antibodies specific for the DNA-binding protein are used to isolate Protein-DNA fragments. After reversal of the crosslink and dissociation of the DNA short read sequencing libraries are prepared [132,133]. Practical considerations: This technique relies on previous knowledge about the DNA-binding proteins and available antibodies. Applications: ChIPseq is commonly used to generate genome wide data on protein-DNA interactions, mainly to determine transcription factor binding sites and their binding dynamics [134]. It has been used also to estimate histone modifications and nucleosome position between different species [72]. Single cell application: [135]
Hi-C*	Summary: DNA-binding proteins and chromatin are covalently crosslinked with formaldehyde and digested with a restriction enzyme. The resulting fragments are ligated to create chimeric molecules of DNA which are further isolated for library preparation and sequencing [136]. Practical considerations: Hi-C relies on restriction enzyme recognition sites which can create bias due to their heterogeneous distribution in the genome [137]. Alternative methods used DNase I [138] or micrococcus digestion [139] to overcome that issue. Applications: Hi-C is commonly used to identify global patterns of 3D genome conformation. Additionally, this method allows exploring how interactions between different chromosomal regions can affect gene regulation. The impact of chromatin topology on gene expression between species has been studied [64,66,140]. Single cell application: [141]
BSseq	Summary: DNA is treated with sodium bisulfite to deaminate cytosine bases into uracil (thymine after PCR) while methyl-cytosine bases are not affected [142]. The treated DNA is then digested for library preparation and sequencing [143]. Practical considerations: The deamination reaction usually has high yield, but small variations can create significant bias in the estimation of global methylation patterns [144]. Since cytosine is converted into thymine, the sequence complexity is reduced, and the strands are no longer complementary causing potential problems with the alignments. However, dedicated software has been developed to deal with the challenging BSseq data analysis (reviewed in [145]). Applications: This method is used to obtain genome wide patterns of DNA methylation which is an important epigenetic modification typically associated with gene expression repression [143]. In recent years, this method has been extensively applied to ecological and evolutionary studies [146,144]. Single cell application: [147,148]