HLA typing sequence alignment TCRshows

How to read the results. ① Identity is the fraction of identical residues or bases within the aligned region. ② Coverage is how much of your sequence was aligned — a partial sequence does not weaken the call, but it does lower coverage. ③ Ranking goes by match count (absolute number of identical positions) first and identity second, so an exact match over a long fragment sorts to the top.
On typing ambiguity. The third and fourth fields of an HLA name (synonymous changes and non-coding regions) are identical at the protein level, and sequencing only exons 2–3 cannot separate many alleles either, so ties for first place are the norm. Equally scoring alleles are merged in the display and summarised as a two-field type such as A*02:01 — the resolution usually reported in the clinic and in the literature.
Which set to use. The protein set (29,380 records / 46,406 alleles) compares amino acid sequences. The nucleotide set is CDS — the coding region after exon splicing (37,700 records / 46,647 alleles) with no introns or UTRs, suited to cDNA or assembled exons. The genomic set (27,847 records, introns and UTRs included) is for comparing raw genomic fragments directly — putting an intron-containing sequence into the CDS set gives very low coverage, and vice versa. A single genomic record can run to 13 kb, so that set is split by locus and loaded on demand; with “all loci” selected, the locus is first determined from representative sequences and only that one is loaded. Note that the genomic set covers far fewer alleles than the CDS set — only those sequenced across the whole gene — so a final type is still normally called from the protein or CDS set.