Atlantic herring GTseq163 panel · 5 fish · 76 bp paired-end reads

GTseq read explorer

Real reads from five fish, followed from raw FASTQ to a genotype. Loci are named by their position in the herring genome (chromosome:start-end). Probes and genotype calls come from the official panel files; extra SNPs found in the reads are shown in gray.

Step 1 · Sort the reads

Where the reads go

Every read starts with the forward primer of the locus it came from, so the first 20 bases tell us which locus a read belongs to. A large share of reads are primer dimers: two primers stuck together with no genomic DNA between them.

Step 2 · Line up reads at one locus

One locus, read by read

Each row is one read pair, merged into one sequence. Teal is the forward primer, violet is the reverse primer, and the colored column is the SNP. Bases shown as dots match the consensus. Red underlined bases are sequencing errors: rare, and scattered rather than lined up in a column.

Step 3 · Count alleles and call a genotype

Probes turn reads into counts

A probe is a short stretch of sequence around the SNP, written once for each allele. The genotyper counts how many reads contain probe 1 and how many contain probe 2, then compares the two counts. A ratio above 10 is called homozygous, a ratio between 0.2 and 5 is called heterozygous, anything in between is left uncalled, and loci with fewer than 10 reads are not called.

Step 4 · Repeat for every locus

Genotypes across the panel

One column per locus (the panel SNP where it was found), one row per fish. Select a cell to open that locus above. This table is the input for estimating heterozygosity, FIS and FST.

Homozygous allele 1 Heterozygous Homozygous allele 2 No call
How this was inferred

Reads were grouped by their first 20 bases (allowing 2 mismatches). Groups with at least 100 reads were treated as loci. Read pairs were merged where R1 and R2 overlap, and adapter sequence was trimmed. Reads shorter than 50 bp that start with a primer were counted as primer dimers.

For each locus, a sequence was counted as a real haplotype if it made up at least 20% of a fish's full-length reads. Positions where real haplotypes differ are SNPs. This page shows the amplicons that carry at least one variable site in these five fish. Another amplified but did not vary, and the rest of the 163-locus panel had too few reads.

Genotypes for the panel SNP are counted the way the GTseq genotyper does it: read 1 must start with the locus's forward primer, then reads containing probe 1 or probe 2 are counted. The primer shown is the first 20 bases of the real primer. Fish 23OB-23 has R1 reads only, so its reads stop at 76 bases.