Article Figures & Data
Figures
- Figure 1
Flow chart showing input files (green), output files (blue), and dependencies (connecting lines) associated with ‘Haplotag’ GBS discovery software. Default file names are shown in yellow, and are normally appended by “.txt” in the Windows file system. Three alternative pipelines (A, B, and C) are available, with required input labeled for each. The cluster discovery pipeline (A), and the haplotype discovery pipeline (B), start by aligning a complete inventory of tags (A), or a reduced inventory of tags from prior work (B), to produce clusters. In (B), the complete inventory is then realigned against this template to increase the sampling of new haplotypes. A complete tag-by-taxa matrix of tag counts (HTBT) is then formed for all tags belonging to clusters of two or more tags. Other output files are then created based on haplotype model fitting. In the production pipeline, only the files labeled by (C) are required, since genotyping is based on counting copies of haplotype-tags in the output files from previous discovery work.
- Figure 2
Passport file produced by Haplotag from simulated demonstration files. Here, six tags (potential haplotypes) are identified at the top. After model fitting by population-based filtering, two locus-models are selected. When Haplotag is run in ‘verbose’ mode, the details of model selection are written in a separate file (see File S2). Locus-1 contains three haplotypes and Locus-2 contains two. SNP positions are identified by color. The table at the bottom of the passport shows the tag counts at the presumed haplotypes within each locus. Counts greater than or equal to one are shaded, indicating that they are scored as “present”.
Tables
- Table 1 Comparison of GBS data analysis using UNEAK vs. Haplotag software
Software Mode Time (hr:min) Number of Loci Passing Population Filter at Indicated Genotype Completeness SNP Loci (80%) Duplicated by Alternate Pipeline Haplotype Loci SNP loci 50% 80% 50% 80% Number Percent UNEAK NA 6:05 NA NA 12,780 4,260 4204a 99%a Haplotag Cluster discovery 6:54b 29,421 11,950 43,378 17,117 4108c 24%c Haplotag Production 0:11b 24,412 7343 31,685 8872 NA NA ↵a Number and percent of SNP-based loci called by UNEAK that were duplicated by Haplotag at the same 80% filtering level.
↵b Times for Haplotag runs do not include the 6 hr required for generation of tag-count files using UNEAK.
↵c Number and percent of SNP-based loci called by Haplotag that were duplicated by UNEAK at the same 80% filtering level.
Supplemental Material for Tinker, Bekele, and Hattori, 2016
Files in this Data Supplement:
- File S1 - Haplotag: Software for Haplotype-Based Genotyping-by-Sequencing (GBS) Analysis User Manual (2016-January-12). (.pdf, 776 KB)
- File S2 - Details of model selection for: HC1. (.zip, 2 KB)
- File S3 - This report shows legacy loci from other work including Huang et al 2014 that match consensus sequences in the current Haplotag production model. (.xlsx, 1.50 KB)
