Genetic analysis of white cell blood groups.
R. Elston
1967-05-01
Abstract:THE IDENTIFICATION of most of the red cell antigens possessed by a particular individual is a relatively simple matter, accomplished by testing a blood sample against a battery of readily available unispecific sera in a series of agglutination reactions. This method of identification, used in population and family studies, has provided the information necessary to unravel the genetic mechanisms involved in the inheritance of such antigens. The identification of white cell antigens, however, has proved to be much more difficult, due to the general unavailability of unispecific sera; hence, an analysis of the genetic mechanisms involved in this case has been seriously hampered. Van Rood (1962) was able to demonstrate the feasibility of establishing genetic groups by using the sera from multiparous women, which, although not unispecific, contain fewer antibodies than sera from multitransfused patients. More recently, a similar approach has been employed by Payne et al. (1964) and by Bodmer and Payne (1965) in the elucidation of additional genetic groups. We shall here be concerned solely with the analysis of data from population studies and shall consider two different types of analyses. The first analysis will be that used by the authors just mentioned, and I shall describe a computer program written to perform such an analysis; the second will be an entirely new approach to the problem, as yet untested, for which we are presently writing computer programs. The data to be analyzed are of the following form. Sera from a sample of n multiparous women are taken, and their reactions with the white cells from each of a panel of m persons are recorded. The agglutination reaction is often scored -, +, + +, + + +, or ++ + +, which we can represent numerically, in order of increasing amount of agglutination, by the numerals 0 through 4. Sometimes, however, intermediate amounts of agglutination also are recognized, so that there are altogether nine possible scores, which we can represent by the numerals 0 through 8. Furthermore, Dr. C. M. Zmijewski, whose data I have been analyzing, performs each agglutination reaction using both undiluted serum and a 1:4 dilution of serum. By adding together the two scores thus obtained, the one from a reaction in which undiluted serum is used and the other from a reaction in which a 1:4 dilution is used, we obtain an integral score, for a given serum with a given cell type, that lies between