Nonrandom sampling in human genetics: Familial correlations

Craig L. Hanis, Ranajit Chakraborty

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

In the context of human genetics, sampling is often nonrandom in that pedigrees are frequently selected by virtue of their having at least one affected individual. For quantitative traits, it may be that the proband is selected because of a phenotypic value above some predetermined cut-point; this will also be true for diseases defined by a cut-point above which individuals are said to be affected (e.g. diabetes, hypertension, and obesity). Relatively little attention has been given to the implications of these forms of nonrandom sampling. In this paper, we show that the biases introduced by such sampling are sufficient to lead to erroneous model specification and biased parameter estimates in path analysis. Existing methodologies used to correct for such sampling (e.g., elimination of proband or regression techniques) also result in similar bias in estimation and model fitting. Thus, it is possible to make inferences which do not reflect underlying biology, but are artifacts of the sampling design. As a consequence, method-of-moment estimators are developed for means, variances, and correlations under such sampling. Simulation results are used to demonstrate the superiority of this method over alternate strategies.

Original languageEnglish
Pages (from-to)193-213
Number of pages21
JournalMathematical Medicine and Biology
Volume1
Issue number2
DOIs
StatePublished - 1 Jan 1984

Fingerprint

Medical Genetics
Sampling
sampling
Pedigree
Artifacts
Obesity
Hypertension
Path Analysis
Moment Estimator
Sampling Design
Model Specification
Diabetes
Model Fitting
Method of Moments
hypertension
path analysis
obesity
diabetes
Alternate
Biology

Cite this

Hanis, Craig L. ; Chakraborty, Ranajit. / Nonrandom sampling in human genetics : Familial correlations. In: Mathematical Medicine and Biology. 1984 ; Vol. 1, No. 2. pp. 193-213.
@article{8bbb67a5d457439b98162ce269ba46e9,
title = "Nonrandom sampling in human genetics: Familial correlations",
abstract = "In the context of human genetics, sampling is often nonrandom in that pedigrees are frequently selected by virtue of their having at least one affected individual. For quantitative traits, it may be that the proband is selected because of a phenotypic value above some predetermined cut-point; this will also be true for diseases defined by a cut-point above which individuals are said to be affected (e.g. diabetes, hypertension, and obesity). Relatively little attention has been given to the implications of these forms of nonrandom sampling. In this paper, we show that the biases introduced by such sampling are sufficient to lead to erroneous model specification and biased parameter estimates in path analysis. Existing methodologies used to correct for such sampling (e.g., elimination of proband or regression techniques) also result in similar bias in estimation and model fitting. Thus, it is possible to make inferences which do not reflect underlying biology, but are artifacts of the sampling design. As a consequence, method-of-moment estimators are developed for means, variances, and correlations under such sampling. Simulation results are used to demonstrate the superiority of this method over alternate strategies.",
author = "Hanis, {Craig L.} and Ranajit Chakraborty",
year = "1984",
month = "1",
day = "1",
doi = "10.1093/imammb/1.2.193",
language = "English",
volume = "1",
pages = "193--213",
journal = "Mathematical Medicine and Biology",
issn = "1477-8599",
publisher = "Oxford University Press",
number = "2",

}

Nonrandom sampling in human genetics : Familial correlations. / Hanis, Craig L.; Chakraborty, Ranajit.

In: Mathematical Medicine and Biology, Vol. 1, No. 2, 01.01.1984, p. 193-213.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Nonrandom sampling in human genetics

T2 - Familial correlations

AU - Hanis, Craig L.

AU - Chakraborty, Ranajit

PY - 1984/1/1

Y1 - 1984/1/1

N2 - In the context of human genetics, sampling is often nonrandom in that pedigrees are frequently selected by virtue of their having at least one affected individual. For quantitative traits, it may be that the proband is selected because of a phenotypic value above some predetermined cut-point; this will also be true for diseases defined by a cut-point above which individuals are said to be affected (e.g. diabetes, hypertension, and obesity). Relatively little attention has been given to the implications of these forms of nonrandom sampling. In this paper, we show that the biases introduced by such sampling are sufficient to lead to erroneous model specification and biased parameter estimates in path analysis. Existing methodologies used to correct for such sampling (e.g., elimination of proband or regression techniques) also result in similar bias in estimation and model fitting. Thus, it is possible to make inferences which do not reflect underlying biology, but are artifacts of the sampling design. As a consequence, method-of-moment estimators are developed for means, variances, and correlations under such sampling. Simulation results are used to demonstrate the superiority of this method over alternate strategies.

AB - In the context of human genetics, sampling is often nonrandom in that pedigrees are frequently selected by virtue of their having at least one affected individual. For quantitative traits, it may be that the proband is selected because of a phenotypic value above some predetermined cut-point; this will also be true for diseases defined by a cut-point above which individuals are said to be affected (e.g. diabetes, hypertension, and obesity). Relatively little attention has been given to the implications of these forms of nonrandom sampling. In this paper, we show that the biases introduced by such sampling are sufficient to lead to erroneous model specification and biased parameter estimates in path analysis. Existing methodologies used to correct for such sampling (e.g., elimination of proband or regression techniques) also result in similar bias in estimation and model fitting. Thus, it is possible to make inferences which do not reflect underlying biology, but are artifacts of the sampling design. As a consequence, method-of-moment estimators are developed for means, variances, and correlations under such sampling. Simulation results are used to demonstrate the superiority of this method over alternate strategies.

UR - http://www.scopus.com/inward/record.url?scp=77957179904&partnerID=8YFLogxK

U2 - 10.1093/imammb/1.2.193

DO - 10.1093/imammb/1.2.193

M3 - Article

C2 - 6600103

AN - SCOPUS:77957179904

VL - 1

SP - 193

EP - 213

JO - Mathematical Medicine and Biology

JF - Mathematical Medicine and Biology

SN - 1477-8599

IS - 2

ER -