Genotyping By Sequencing (GBS)
We are using GBS in many species where a reference genome may not be available. The methods have been published (Morris et al 2011, Elshire et al, 2011, Grabowski et al, 2014). We are using TASSEL GBS UNEAK to align reads and call SNPs denovo. In model organisms (Arabidopsis and Brachypodium) we map reads to the reference genome to get the physical position of the SNPs.
The current protocol uses Pst1 (6 cutter) for reduced representation with 96 barcodes on the forward read and up to 12 barcodes on the reverser read. We are often pooling more than 4 plates per Illumina lane (200M paired end reads). This service is available at ANU Biomedical Resource Facility (~$30/sample). We generally get 20-60k SNPs with 50% missing data.
We have also used Hpa2/Msp1 (CCGG) with 24*24 dual end barcodes = 576 sample combinations. Although read counts per sample per locus (>80k) are low population and family structure is still obvious, as low as $10/sample.
Whole Genome Sequencing
Our favoured approach is Nextera, which uses tagmentation to insert adaptors at largely random locations. Multiplexing is performed by using indexed primers at the enrichment PCR step. A reduced representation of whole genome library is possible via NextRad.
Whole Genome Bisulfite Sequencing (WGBS)
The profiling of cytosine methylation across the genome is largely done using Illumina TruSeq Methylation kits (formally Epignome from Epicentre) based on post-bisulfite adapter ligation (PBAT). Profiling has also performed by Pre-Bisulfite adapter ligation (Lister) method using methylated barcode oligonucleotides.
Transcriptome analysis can be performed by first converting RNA to cDNA. Then Nexterra tagmentation add adaptors and barcodes for multiplexing reducing cost. Alternatively others have had success with high throughput RNAseq.