Tassel

Presence/absence scores for each tag were used in a binomial test of segregation versus an independent framework map. For maize, this framework map consisted of 644 SNPs genetically mapped in the maize nested association mapping (NAM) population [29] (link) and then genotyped in the IBM population. The binomial segregation test filtered for sequence tags that co-segregated with only one of the two parental alleles at a given SNP. For each SNP marker, the two possible parental sources of a tag were each tested in turn. A “success” was recorded when a tag co-occurred in a RIL with the SNP allele from its presumed parental source, otherwise a “failure” was recorded. The binomial sample size was the number of RILs in which the tag was present and the SNP was not missing or heterozygous. For maize, tests were only performed if the sample size was at least 10. The probability of success was defined as the proportion of the RILs that contained the SNP allele being tested. For maize, a threshold p-value of 0.001 was considered significant for directed tests versus the physically closest SNP, or 0.0001 for elsewhere in the genome.
For barley, mapping was conducted using flanking SNPs and a threshold of p<0.0001 for the binomial test. In practice, a sequence tag was mapped in barley only if it always co-occurred with one SNP allele and never the other.
In maize only, biallelic GBS markers were identified as follows. Pairs of tags that aligned to exactly the same unique position and strand in the maize reference genome (B73 RefGen v1) and that also co-segregated with the physically closest SNP (p<0.001) were merged into a single, biallelic marker. These markers were then re-tested for co-segregation with the physically closest SNP using Fisher's Exact Test (p<0.001). Biallelic GBS markers that passed the latter test were then incorporated into a high density, framework map and ordered according to their positions in the reference genome. To determine how many of the remaining presence/absence GBS tags could be genetically mapped in maize, the binomial test of segregation was repeated versus this high density framework map, with a threshold of p<0.0001.
Software for the sequence filtering and the mapping analysis was written in Java and is available on SourceForge (http://sourceforge.net/projects/tassel/). This software is part of the TASSEL package but is not currently implemented in the TASSEL GUI.

We collected GBS data from a collection of 1995 accessions from the genus Malus from the US Department of Agriculture apple germplasm repository in Geneva, NY. The samples were processed with two different restriction enzymes (ApeKI, PstI/EcoT22I) in separate GBS libraries and were sequenced using Illumina Hi-Sequation 2000 technology. Genotypes were called using a custom GBS pipeline described in Gardner et al. (2014) (link). Briefly, 100-bp reads generated from both enzymes were aligned to the Malus domestica reference genome version 1.0 (Velasco et al. 2010 (link)) using the default parameters in BWA (Li and Durbin 2009 (link)). Genotypes were called using GATK (McKenna et al. 2010 (link)) with a minimum of eight reads supporting each genotype. The final genotype matrix was filtered to contain only samples from the domesticated apple, Malus domestica, and ≤20% missing data per SNP and per sample. SNPs with a minor allele frequency (MAF) of <0.01 were then discarded. Finally, the data were pruned to exclude clonal relationships: if two or more samples had IBD >0.9, they were considered clones and the sample with the least amount of missing data from the group was retained. This resulted in a dataset of 711 samples and 8404 SNPs.
To test the accuracy of our imputation method we created a “masked” dataset by setting 10,000 random genotypes to missing. This created “truth known” genotypes to which our imputed genotype calls were compared. We limited our testing to 10,000 masked genotypes, which represents 0.17% of the genotype matrix, in order to maintain a dataset with a reasonable amount of missing data while providing enough masked genotypes to be able to estimate imputation accuracy.
Biased allele frequency in imputed data has been shown to affect downstream analyses (Han et al. 2014 (link)). To determine how well each imputation method estimates allele frequencies, we filtered the genotype matrix to contain no missing data. This resulted in a matrix containing 1001 SNPs from 459 samples (Figure S2). We masked and then imputed 20% (91,952 genotypes) of the genotypes at random and compared the allele frequency estimates from the imputed data to the allele frequency estimates from the complete genotype matrix. As most imputation methods make use of other SNPs to aid imputation, we imputed using all 8404 SNPs in the dataset so as to provide more information to these methods. We then restrict our analysis to the 1001 complete SNPs.
We also tested the performance of our method on genome-wide SNP data from maize and grape. The maize data were downloaded from the International Maize and Wheat Improvement Center (Hearne et al. 2014 ). We reduced the data to biallelic SNPs with <20% missing data and a MAF >1% and then discarded samples with >20% missing data. This resulted in 43,696 SNPs from 4300 samples.
To generate the grape dataset we collected GBS data from a collection of diverse samples from the genus Vitis including commercial Vitis vinifera varieties, hybrids and wild accessions from the USDA grape germplasm collection. The samples were processed with two different restriction enzymes (HindIII/BfaI, HindIII/MseI) and were sequenced using Illumina Hi-Sequation 2000 technology. We then used the 12X grape reference genome (Jaillon et al. 2007 (link); Adam-Blondon et al. 2011 ) and the Tassel / BWA version 4 pipeline to generate a genotype matrix (Li and Durbin 2009 (link); Glaubitz et al. 2014 (link)). Default parameters were used at each stage except for the SNP output stage where we filtered for biallelic SNPs. We then removed any genotypes with fewer than eight supporting reads using vcftools (Danecek et al. 2011 (link)). Using PLINK (Purcell et al. 2007 (link)), we removed SNPs with >20% missing data before removing samples with >20% missing data. We then removed SNPs with excess heterozygosity (failed a Hardy−Weinberg equilibrium test with a p-value < 0.001) and finally SNPs with a MAF < 0.01. This created a dataset of 8506 SNPs and 77 samples.

GBS data analysis was performed using the GBS discovery pipeline of TASSEL version 5.0 software [30 (link)]. The FASTQ and sample key files (containing the barcodes for each genotype) generated from raw sequence reads by the CASAVA 1.8.2 software package (Illumina Inc.) were used as input for processing in the pipeline. Before the analysis, 64-base reads were generated by trimming reads having the barcodes for each genotype followed by an ApeKI cut site using the FastqToTagCountPlugin of the pipeline. Reads with unidentified bases (N) were excluded from analysis. The barcoded sequence reads were collapsed into unique sequence tags with counts using the FastqToTagCountPlugin with default parameters with the exception that minimum number of times a tag must be present was set to 3. Tag count files that contained the sequence tags that passed the minimum count threshold of 3 were merged into a master file using the MergeMultipleTagCountPlugin. The master tags in FASTQ format generated by TagCountToFastqPlugin were aligned to the tomato S. lycopersicum reference genome using the bowtie2 plugin with default parameters [31 , 32 (link)]. SAMConverterPlugin generated the “Tags On Physical Map” (TOPM) file which contained information about the physical positions of the master tags which had the best unique alignments with the reference genome. In addition to the TOPM file, the “Tags by Taxa” (TBT) file that contained tag counts of each barcode generated by FastqToTBTPlugin was used for SNP calling according to the parameters of the TagsToSNPByAlignmentPlugin (Additional file 1: Table S1). SNPs were recorded in a HapMap file for each chromosome. MergeDuplicateSNPsPlugin was used to merge the duplicate SNPs. SNPs were filtered based on minimum Taxon Coverage (mnTCov: 0.01), minimum Site Coverage (mnSCov: 0.2), linkage disequilibrium with neighboring SNPs (hLD: TRUE), minimum R2 value for the LD filter [−mnR2]: 0.2, and minimum Bonferroni-corrected p-value for the LD filter [−mnBonP]: 0.005. A physical map of the identified SNPs was drawn using Mapchart software [33 (link)].

Leaf samples were collected for sequencing from 52 individuals within the progeny population. Since the identification of outliers is the most critical goal of genomic selection, a sampling method was used increase the incorporation of outlier individuals in the validation set. Specifically, individuals were sampled using weights from an inverted density distribution of the population's mean Z-scores. The mean Z-scores were calculated from each individual's heading date and winter survivorship scores. This resulted in a subset of the population with trait values slightly oversampled from the tails of the gaussian distribution. Genotyping by sequencing occurred on an Illumina sequencer (NovaSeq 6000) through the University of Wisconsin Biotechnology Center using PstI-MspI restriction enzyme digestion before ligating fragments to barcoded adaptors prior to polymerase chain reaction amplification. Data analysis of sequencer output used TASSEL (Glaubitz et al. 2014 (link)). Briefly, the barcoded sequence read outputs were collapsed into a set of unique sequence tags with counts. Tags were aligned to the reference genome (P. virgatum v5.1), assigning each tag to a position with the best unique alignment. The occupancies of tags for each sample were observed from barcode data. Resulting files were used to call single-nucleotide polymorphism markers (SNPs) at the tag locations on the genome, resulting 1,072,642 SNPs.

Meng et al. (2017) (link) genotyped the MAGIC population with a 55K SNPs array. Selection of high-quality SNPs for QTL mapping used a three-step filtering strategy. First, markers monomorphic among the four parents were removed. Second, set all heterozygous genotypes to “deletion” and delete markers with deletion values greater than 10%. Finally, markers with a minor allele frequency of less than 3% were deleted. The number of markers remaining was 22,160.
The MLM (Mixed Linear Model) implemented in TASSEL version 5.2.3 was used to analyze the associations between SNP markers and traits. P < 0.001 was used as the threshold to declare the significance of marker-trait associations. R² was used to evaluate the percentage of phenotypic variance explained of related loci to phenotypic traits.

In total 431, 816, 1491, and 1029 advanced breeding lines of F₇-F₉ generations along with five released varieties (BRRI dhan28, BRRI dhan29, BRRI dhan67, BRRI dhan74, and BRRI dhan89, were evaluated for yield at multi-locations during Boro season of 2018–19, 2019–20, 2020–21, and 2021–22, respectively. The trial meta-data can be seen the Supplementary Table S1. Green leaf tissues from a representative plant of each breeding line was collected in labeled glassine bag at 4–5 weeks after transplanting and stored immediately on ice. The samples were stored in a −80°C freezer until processing for genotyping. DNA was isolated and purified according to the modified Cetyltrimmethyl Ammonium Bromide (CTAB) protocol (Aboul-Maaty and Oraby 2019 (link)). Genotyping with genome-wide 1024 SNP markers including 92 trait-specific markers named as 1K-RiCA panel (Arbelaez et al., 2019 (link)) was performed at an outsourcing genotyping service provider with the help of IRRI Genotyping Services Laboratory, The Philippines. The genotyping data of 1k-RiCA SNPs were filtered using TASSEL v5.0 (Bradbury et al., 2007 (link)) following the criteria that the individuals with more than 15% of heterozygous loci were removed, markers with more than 15% of missing values and minor allele frequency below 0.05 were removed. After filtering, 814–889 markers were retained for doing downstream analysis.