SNP genotyping in Pseudotsuga menziesii and Pinus radiata using targeted genotyping-by-sequencing (GBS): Improved Bayesian SNP calling using a beta-binomial distribution and other optimized input parameters

Documentation for: SNP genotyping in Pseudotsuga menziesii and Pinus radiata using targeted genotyping-by-sequencing (GBS): Improved Bayesian SNP calling using a beta-binomial distribution and other optimized input parameters Background: Single-nucleotide polymorphism markers (SNPs) have important applications in gene conservation, breeding, and fundamental genetics research. Our long-term goal is to develop routine approaches for SNP genotyping in forest trees. Ideally, these approaches would be inexpensive, able to accommodate a wide range of samples and SNPs, available through commercial providers, and produce high-quality SNP data.Results: Using targeted genotyping-by-sequencing (GBS), we developed SNP assays for two highly heterozygous tree species, Douglas-fir (Pseudotsuga menziesii) and radiata pine (Pinus radiata). Using Douglas-fir haploid and diploid data, we optimized Bayesian SNP calling by testing four input parameters: (1) allele and genotype prior probabilities, (2) Rho, the beta-binomial dispersion parameter, (3) estimated read error (BayesReadError), and (4) the logPO cutoff used to filter low confidence SNP calls. logPO is the Bayesian posterior odds ratio for a called SNP. Compared to assuming a binomial distribution of read counts (Rho = 0), the beta-binomial distribution (Rho = 0.33) substantially reduced call error and heterozygote undercalling. For Douglas-fir, we tested 5,360 SNP assays, and then studied the performance of the best 4,000. For radiata pine, we tested 6,000 SNP assays, and then studied the performance of the best 4,565. In Douglas-fir and radiata pine, our Bayesian approach resulted in median call rates of 95% to 98% for the top-ranked SNPs, with an estimated call error of 1.60% for known homozygous genotypes and 2.27% for known heterozygotes. In radiata pine, median and mean call rates were above 91% for GBS and genotyping using a fixed Axiom genotyping array. Additionally, the median correspondence between the GBS and Axiom genotypes was about 98% overall (mean 96%).Conclusions: By optimizing Bayesian SNP calling, selecting the best 4-5K SNPs, and excluding samples with low DNA amounts, we substantially reduced call error and heterozygote undercalling, resulting in SNP genotypes that were nearly identical to genotypes obtained using the Axiom array. Furthermore, genotyping performance should increase even further if our SNP rankings were used to develop less complex probe pools that target fewer SNPs.NCBI BioSamples: https://www.ncbi.nlm.nih.gov/biosample SAMN37626997-SAMN37627956: Douglas-fir control-cross radicle and megagametophyte tissues. SAMN37622077-SAMN37622610 and SAMN37626400-SAMN37626944: Douglas-fir open-pollinated leaf tissues. SAMN37666260-SAMN37666311: Radiata pine leaf tissues. Raw sequences: Raw sequence reads are archived under NCBI BioProjects https://www.ncbi.nlm.nih.gov/bioproject/. PRJNA1022809: Sequence reads for Douglas-fir control-cross (CC) samples. PRJNA1022573: Sequence reads for Douglas-fir open-pollinated (OP) samples. PRJNA1023615: Sequence reads for radiata pine samples. File names: Douglas-fir_CC_calls_2023-11-07.txt Douglas-fir_OP_calls_2023‑11‑07.txt Description: SNP calls for individual Douglas-fir control-cross (CC) or open-pollinated (OP) samples using targeted GBS and optimized Bayesian SNP calling. Bayesian parameters were SNP.FLAT priors, BayesReadError = 0.005, and Rho = 0.33. Variables: Affx_snpID: Axiom SNP ID previously assigned by Affymetrix/Thermo Fisher Scientific (Howe et al 2020). AX_probeset_ID: Axiom probeset_ID previously assigned by Affymetrix/Thermo Fisher Scientific(Howe et al 2020). sample_name: Organization_project_plate_well of the DNA sample in a 96-well plate (OSU = Oregon State University). family: Family numbers for the Douglas-fir control-cross samples. tissue: R = radicle/root, M = megagametophyte, F = foliage. A, C, G, T: Read depth (counts) for the A, C, G, and T nucleotides. call: Diploid call for the R and F tissues or haploid call for the M tissues with nucleotides listed in alphabetical order and NA denoting a missing value (no-call). logPO: Log of the Bayesian posterior odds ratio for the called diploid or haploid genotype. File name: Radiata_pine_calls__2023-11-07.txt Description: SNP calls for individual radiata pine samples using targeted GBS and optimized Bayesian SNP calling. Bayesian parameters were SNP.FLAT priors, BayesReadError = 0.005, and Rho = 0.33. Variables: Affx_snpID: Axiom SNP ID previously assigned by Affymetrix/Thermo Fisher Scientific (Graham et al. 2022). AX_probeset_ID: Axiom probeset_ID previously assigned by Affymetrix/Thermo Fisher Scientific (Graham et al. 2022). sample_name: Organization_project_plate_well of the DNA sample in a 96-well plate (SCI = Scion New Zealand Forest Research Institute). tissue: F = foliage. A, C, G, T: Read depth (counts) for the A, C, G, and T nucleotides. call: Diploid call for the radiata pine sample with nucleotides listed in alphabetical order and NA denoting a missing value (no-call). logPO: Log of the Bayesian posterior odds ratio for the called diploid genotype. File name: Douglas-fir_targeted_GBS_snps_2023-11-07.txt Description: Summary information for SNPs called using targeted GBS and optimized Bayesian SNP calling. Bayesian parameters were SNP.FLAT priors, BayesReadError = 0.005, Rho = 0.33, and logPO_cutoff = 0 or 1. SNPs were summarized based on calls made from diploid control-cross (CC) and diploid open-pollinated (OP) samples. Variables: snpRank: Final rank for Douglas-fir SNPs by combining the ranks from the CC and OP samples. The ranks for the CC and OP samples were based on median logPO with logPO_cutoff = 0. Affx_snpID: Axiom SNP ID previously assigned by Affymetrix/Thermo Fisher Scientific (Howe et al 2020). AX_probeset_ID: Axiom probeset_ID previously assigned by Affymetrix/Thermo Fisher Scientific (Howe et al 2020). snpID: SNP information based on the original transcriptome sequencing and assembly (Howe et al 2013). scID: Scaffold ID for the SNP from the Douglas-fir reference sequence database v1.0 (https://treegenesdb.org/FTP/Genomes/Psme/). sc_start: Scaffold start position for the SNP from the Douglas-fir reference sequence database v1.0 (https://treegenesdb.org/FTP/Genomes/Psme/). sc_end: Scaffold end position for the SNP from the Douglas-fir reference sequence database v1.0 (https://treegenesdb.org/FTP/Genomes/Psme/). snp: Inferred SNP type with nucleotides listed in alphabetical order. CR: Call rate for unfiltered calls (logPO_cutoff = 0). Values are averages for the CC and OP samples. CR_f: Call rate for filtered calls (logPO_cutoff = 1). Values are averages for the CC and OP samples. logPO: Median logPO for unfiltered calls (logPO_cutoff = 0). Values are averages for the CC and OP samples. logPO_f: Median logPO for filtered calls (logPO_cutoff = 1). Values are averages for the CC and OP samples. File name: Radiata_pine_targeted_GBS_snps_2023-11-07.txt Description: Summary information for SNPs called using targeted GBS and optimized Bayesian SNP calling. Bayesian parameters were SNP.FLAT priors, BayesReadError = 0.005, Rho = 0.33, and logPO_cutoff = 0 or 1. SNPs were summarized based on calls made from diploid samples. Variables: snpRank: Final rank for radiata pine SNPs based on median logPO with logPO_cutoff = 0. Affx_snpID: Axiom SNP ID previously assigned by Affymetrix/Thermo Fisher Scientific (Graham et al. 2022). AX_probeset_ID: Axiom probeset_ID previously assigned by Affymetrix/Thermo Fisher Scientific (Graham et al. 2022). scID: Scaffold ID for the SNP from an unpublished radiata pine reference genome sequence (Graham et al. 2022). sc_start: Scaffold start position for the SNP from an unpublished radiata pine reference genome sequence (Graham et al. 2022). sc_end: Scaffold end position for the SNP from an unpublished radiata pine reference genome sequence (Graham et al. 2022). snp: Inferred SNP type with nucleotides listed in alphabetical order. CR: Call rate for unfiltered calls (logPO_cutoff = 0). CR_f: Call rate for filtered calls (logPO_cutoff = 1). logPO: Median logPO for unfiltered calls (logPO_cutoff = 0). logPO_f: Median logPO for filtered calls (logPO_cutoff = 1). Citations: Graham, N., Telfer, E., Frickey, T., Slavov, G., Ismael, A., Klápště, J., and Dungey, H. 2022. Development and validation of a 36K SNP array for radiata pine (Pinus radiata D.Don). Forests 13(2):176. Howe, G.T., Jayawickrama, K., Kolpak, S.E., Kling, J., Trappe, M., Hipkins, V., Ye, T., Guida, S., Cronn, R., Cushman, S.A., and McEvoy, S. 2020. An Axiom SNP genotyping array for Douglas-fir. BMC Genomics 21(9). DOI: https://doi.org/10.1186/s12864-019-6383-9 Howe, G.T., Yu, J., Knaus, B., Cronn, R., Kolpak, S., Dolan, P., Lorenz, W.W., and Dean, J.F.D. 2013. A SNP resource for Douglas-fir: de novo transcriptome assembly and SNP detection and validation. BMC Genomics. 2013 Feb 28;14:137. doi: 10.1186/1471-2164-14-137. PMID: 23445355; PMCID: PMC3673906.