Linked Life Data

15545394

Source:http://linkedlifedata.com/resource/pubmed/id/15545394

Statements in which the resource exists as a subject.
PredicateObject
rdf:type
lifeskim:mentions
pubmed:issue
12
pubmed:dateCreated
2004-11-16
pubmed:abstractText
The objective of this work was to propose an algorithm (HAPROB) to estimate haplotype probabilities for genotyped members of half-sib families for which parents lacked genotypic information. The algorithm had 2 basic steps. First, a Monte Carlo-based approach was used to estimate haplotype probabilities for sires conditional upon offspring genotypes and population allelic frequencies, and then offspring-haplotype probabilities were estimated conditional upon sire probabilities and population frequencies. The 2 steps were alternated iteratively until estimates of population frequencies were essentially unchanged. Simulation was used to evaluate effects of the number of Monte Carlo cycles on the accuracy of the reconstructed haplotypes. Fifty thousand cycles was found to be sufficient for the haplotype configurations considered. Accuracy of the algorithm was compared with that obtained by the public domain SIMWALK2 software. Predictions of the most likely haplotype configurations are produced by SIM-WALK2, but no estimates of probability are given. The accuracy of the current approach was comparable to that obtained from SIMWALK2. The proportions of times that haplotypes were reconstructed correctly were 87.0 and 92.4% (sires and offspring) for HAPROB vs. 87.5 and 91.5% for SIMWALK2. Effects of family size on accuracy of reconstruction were examined. Accuracy of reconstruction was only about 4% for sires with 2 offspring, but accuracy among the offspring themselves was 65%. Accuracy increased quickly as family size increased and reached 100% for sires with 30 offspring. Maximum accuracy for offspring was about 96%. Estimates of haplotype probabilities produced can be used in regression analyses to estimate effects of haplotypes on quantitative phenotypes.
pubmed:language
eng
pubmed:journal
pubmed:citationSubset
IM
pubmed:status
MEDLINE
pubmed:month
Dec
pubmed:issn
0022-0302
pubmed:author
pubmed:issnType
Print
pubmed:volume
87
pubmed:owner
NLM
pubmed:authorsComplete
Y
pubmed:pagination
4303-10
pubmed:dateRevised
2010-11-18
pubmed:meshHeading
pubmed:year
2004
pubmed:articleTitle
A Monte Carlo approach for estimation of haplotype probabilities in half-sib families.
pubmed:affiliation
Istituto di Biologia e Biotecnologia Agraria, Consiglio Nazionale delle Ricerche, Segrate 20090, Italy. boettch@ibba.cnr.it
pubmed:publicationType
Journal Article, Research Support, Non-U.S. Gov't