Selection of a Set of High-Quality Data for B Cell Antigen Analysis

Abstract

In order to obtain the best definition of splits in the DR system and to facilitate analysis of the “DR” monoclonal antibodies, a file of high-quality B cell typing data was required. The aim was to select labs that had good B cell typing and, within the data of those labs, to select cells with a clear-cut separation of positive and negative scores, i.e., data in which the scores were clearly distributed between 1 and 6/8, with only a few scores of 2 or 4. Data that did not show clear bimodality were considered unsatisfactory because of the blurred distinction between positive and negative antibody reactions.

We acknowledge with gratitude the financial support from Health and Welfare Canada (Visiting Scientist Award to W.H.M.); D.F.G grant number AL 92/13–1, 13–2 (to E.D.A.) and from the Deutscher Akademischer Austauschdienst (grant as a Visiting Scientist to T.B.P.).