Serogroup or antigen

Averaging a group of beads into one antibody, or splitting out the antigen that carries the excess.

Chapter 2 — Serogroup or antigen

Averaging vs. splitting

When several beads carry the same serological group, SAgA must decide how finely to report the antibody: as a single group-level call, or split into an antigen-specific one. It compares the beads' MFI.

Situation	What SAgA does
MFI homogeneous — values close, within usual bead variability	Averages them into a single serological-group antibody.
MFI heterogeneous — one bead stands far above the rest	Splits: a group antibody shared by all beads, plus an antigen-specific one carrying the excess.

“Heterogeneous” depends on how many beads carry the serogroup: with two beads, one standing far above the other; with more than two, at least one standing above the rest.

Homogeneous · average into one serogroup antibody

Three A2 beads with close MFI around 9000 are averaged into a single A2 antibody at 9285 — The three A2 beads read 8000, 10655 and 9200 — close enough to be bead-to-bead variability. SAgA averages them into a single A2 antibody at 9285.

Heterogeneous · split off the antigen carrying the excess

Three A2 beads where A*02:03 at 12545 stands far above the others — Now the same three beads read 1677, 12545 and 2104 — one stands far above the others. The gap is too large to be mere bead variability.

SAgA splits into a broad A2 antibody at 1890 plus an antigen-specific A*02:03 at 10655 — So SAgA splits the reactivity: a broad A2 antibody at 1890 shared by all three beads, plus an antigen-specific antibody against A*02:03 carrying the excess (10655).

Several alleles, several readings

With more alleles the picture is richer — say A2, A*02:03 and A*02:06 all reactive, in various patterns (all homogeneous; A2 alone plus a split A*02:03/06; or three distinct levels). Again, the user decides the strategy: are there several antibodies, each specific for one allele, or fewer antibodies specific for several alleles? It may equally be a matter of differential affinity — a single antibody raised against the three alleles, binding each to a different degree.

The choice follows a smooth rule: the wider the gap between beads, the more SAgA leans towards splitting off a separate antigen-specific antibody. The default transition is calibrated from the combined experience of senior HLA experts, so SAgA's behaviour matches what an expert would conclude.

A smooth curve relating the MFI gap between beads to the tendency to split — Averaging when MFI are close, splitting when the gap is large — a smooth, expert-calibrated transition rather than a hard cut.

Cases for this part

A panel — one serogroup averaged, one split31 beads