The use of computers

Abstract

The use of the computer in linkage in man has been limited to the basic two-locus incomplete-information problem, and little consideration has been given to the implications, or of the limitations of two-locus analysis. The computer must be seen not only as an aid to the storage and interpretation of raw data, but also as a competitor to the funding of more raw data, and as an obstacle to the development of new methods of analysis through the misapplication of programs devised for other purposes. In general, data must be transcribed in a suitable form, usually with a necessary loss of information, and then this transcribed form must be translated in a way which improves understanding, as by graphical output, or leads to additive results which can be used to combine data from various sources, as in the z-lods of MORTON. The transcription and translation are usually associated with loss of information, and also with a loss of 'immediacy,' that is, of the symbols or numbers used being inadequately exploited to maximize concept formation. The concept of 'immediacy,' and its exploitation as an aim of a coding system, has not been appreciated. The computer, whose internal representation is so similar to that of both the genetic message and those manifestations normally coded in laboratory workbooks as series of +'s and -'s, such as ++-+ for an MNSs blood group phenotype, can impose a useful discipline on coding systems without any loss of the aesthetic advantages of a full set of conventional characters, including the lower case letters advocated by MENDEL. Unless coding systems are sympathetic to the demands of the computer, and retain immediacy in meaning and fidelity in mediation between workbook and core-image, the workings of the computer are likely to be even more mysterious to the laboratory worker.