Genetics and the epidemiology of chronic diseases : a symposium, June 17-19, 1963, Ann Arbor, Michigan ... / edited by James V. Neel, Margery Shaw, William J. Schull.

Date:
1965
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: Genetics and the epidemiology of chronic diseases : a symposium, June 17-19, 1963, Ann Arbor, Michigan ... / edited by James V. Neel, Margery Shaw, William J. Schull. Source: Wellcome Collection.

    18/412 (page 8)
    In addition to true gene dominance, we may speak of trait domi¬ nance. This will appear for traits for which the proper tools for heterozygote recognition are not available to the investigator. For our discussion today we will speak of dominant and recessive traits rather than dominant and recessive genes since the trait is the ob¬ served event from which the investigation of etiology proceeds. The problems of ascertainment will be discussed in detail by Dr. Crow later in this program. We shall only note here that family sampling is complicated by dominance relationships, by the small size of human famihes, which permit some parents potentially capa¬ ble of producing affected offspring to be omitted from a sample, and by differing probabilities of including in a sample families having more than one affected offspring. Family samples collected under various types of ascertainment must thus be treated statistically in different manners to obtain comparable геИаЫе results. The investigator is generally unable to measure directly the action of a single gene locus. He is required to work with end-products, produced through multiple developmental and physiologic inter¬ actions during the lifetimes of the individuals. Environmental factors or other genes segregating in a family may produce variation in the expression of a genetic trait or may prevent the appearance of symp¬ toms in a genetically susceptible individual (incomplete penetrance). Variable expression presents no problem for genetic analysis if the different manifestations are recognizable, but incomplete penetrance, since it leads to a modification of the observed ratio among the offspring, is a serious problem. Genetic analysis is impossible if penetrance is low. If a disease having delayed onset is rare, the penetrance may be calculated from the distribution of age of onset in family members, without any genetic assumptions (Morton, 1957). If incomplete penetrance is entirely due to delayed onset, the average penetrance lies between УУ(г) 6^(г)с?г and y/i (2) 6'(г)с?2 where/(2) is the frequency of age z at death or last examination among normal and affected siblings, /1(2) is the same frequency with the index cases excluded, and G{2) is the cumulative frequency of onset at age z among affected cases. Heterogeneity in age of onset among families will cause such a penetrance estimate to be approximate only (Morton, 1957). For a dominant trait which occasionally skips a generation, pene¬ trance may be estimated from the number of unaffected carriers in the direct line of descent from an affected ancestor to the first affected descendant in that line, if the trait is sufficiently rare that the chance of a normal carrier mating into the family is negligible. If the penetrance is low, however, long skips may fail to be recorded so that the penetrance will be overestimated. A proportion of the cases observed in a body of data may have arisen through nonrandom nonrecurrent events such as mutation, 10