Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Apr 23;3(4):e1977.
doi: 10.1371/journal.pone.0001977.

Genome Reshuffling for Advanced Intercross Permutation (GRAIP): Simulation and Permutation for Advanced Intercross Population Analysis

Affiliations
Free PMC article

Genome Reshuffling for Advanced Intercross Permutation (GRAIP): Simulation and Permutation for Advanced Intercross Population Analysis

Jeremy L Peirce et al. PLoS One. .
Free PMC article

Abstract

Background: Advanced intercross lines (AIL) are segregating populations created using a multi-generation breeding protocol for fine mapping complex trait loci (QTL) in mice and other organisms. Applying QTL mapping methods for intercross and backcross populations, often followed by naïve permutation of individuals and phenotypes, does not account for the effect of AIL family structure in which final generations have been expanded and leads to inappropriately low significance thresholds. The critical problem with naïve mapping approaches in AIL populations is that the individual is not an exchangeable unit.

Methodology/principal findings: The effect of family structure has immediate implications for the optimal AIL creation (many crosses, few animals per cross, and population expansion before the final generation) and we discuss these and the utility of AIL populations for QTL fine mapping. We also describe Genome Reshuffling for Advanced Intercross Permutation, (GRAIP) a method for analyzing AIL data that accounts for family structure. GRAIP permutes a more interchangeable unit in the final generation crosses - the parental genome - and simulating regeneration of a permuted AIL population based on exchanged parental identities. GRAIP determines appropriate genome-wide significance thresholds and locus-specific P-values for AILs and other populations with similar family structures. We contrast GRAIP with naïve permutation using a large densely genotyped mouse AIL population (1333 individuals from 32 crosses). A naïve permutation using coat color as a model phenotype demonstrates high false-positive locus identification and uncertain significance levels, which are corrected using GRAIP. GRAIP also detects an established hippocampus weight locus and a new locus, Hipp9a.

Conclusions and significance: GRAIP determines appropriate genome-wide significance thresholds and locus-specific P-values for AILs and other populations with similar family structures. The effect of family structure has immediate implications for the optimal AIL creation and we discuss these and the utility of AIL populations.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. RI and AIL breeding schemes.
The left panel of this figure diagrams breeding of a small recombinant inbred (RI) strain set. Each strain is essentially a set of repeated intercrosses starting with inbred parental strains. The animals are considered inbred at 20 generations. The right panel is an example of breeding an advanced intercross line. (Typically such lines consist of 50–100 animals at each generation rather than the 8 shown.) Letters A–H indicate 8 unique F2 animals, and their offspring “inherit” these identifiers. Breeding pairs are chosen for minimum relatedness at each generation.
Figure 2
Figure 2. The GRAIP approach.
This cartoon summarizes the GRAIP approach. First (1) parental happlotypes are regenerated if they are not already known and (2) the parents are permuted. Next (3) the population of offspring is regenerated using the permuted genotypes and (4) permuted maps are generated using the non-permuted phenotypes. Finally, (5) the significance of the permuted maps are compared at a pointwise and whole genome basis with the original map.
Figure 3
Figure 3. Coat color (A) and hippocampus weight (B) in the UTHSC AIL population Red traces are the simple mapping output, and the red bar is genome-wide P = 0.05 by naïve permutation.
Black traces are GRAIP permutation output. Note that for ease of graphing on a -log scale we have adjusted P<1/10000 to P = 0.0001, so the maximum –log P = 4. Simple mapping results are on the left hand scale, while GRAIP results are on the right. On the Chr.4 coat color locus simple mapping value is truncated at LOD = 25, to simplify reading the graph. Shaded gray regions are significant at genome-wide P = 0.05 or better in the GRAIP results.
Figure 4
Figure 4. BXD coat color QTL-maps.
Coat color QTL maps treating BXD observations as independent individuals versus mapping strain means. (A) Comparison of simple mapping and GRAIP for BXDs treated as individuals. Red traces are simple mapping output, and the red bar is P = 0.05 for the naïve permutation. Black traces are GRAIP mapping output (5000 permutations) and shaded gray region is significant at genome-wide 0.05 or better in the GRAIP results. (B) simple mapping output for BXD strain means. Black bar indicates P = 0.05 for the naïve (appropriate, in this case) permutation.
Figure 5
Figure 5. Box plot of hippocampus weight by family.
Whiskers represent the distribution of the highest and lowest 25% of observations. The line across the box represents the median value, while the “+” indicates the mean. Family 2 is missing because there were not hippocampus weight observations in that group.
Figure 6
Figure 6. Effects of progressively removing samples from a population.
Measurement were taken at the position with the best P-value in the Hipp1a locus. Samples removed either by family or by an equivalent number of randomly selected individuals. –log P measured at the most significant position in the original BXD data set for hippocampus weight, near the physical center of the Hipp1a interval.
Figure 7
Figure 7. Variation in LOD score distribution by position on Chr 1.
Distribution of 95th percentile LOD scores by marker for 10,000 GRAIP permutations of coat color and body weight QTL mapping in the AIL population. Note that the maximum and minimum values of the 95th percentile on this chromosome alone are separated by a difference in LOD of 8.3 for coat color and 1.3 for body weight, which indicates that the same LOD score is equivalent to a considerably different P value depending on position and original phenotype.

Similar articles

See all similar articles

Cited by 24 articles

See all "Cited by" articles

References

    1. Korstanje R, Paigen B. From QTL to gene: the harvest begins. Nat Genet. 2002;31:235–236. - PubMed
    1. Abiola O, Angel JM, Avner P, Bachmanov AA, Belknap JK, et al. The nature and identification of quantitative trait loci: a community's view. Nat Rev Genet. 2003;4:911–916. - PMC - PubMed
    1. Darvasi A, Soller M. Advanced intercross lines, an experimental population for fine genetic mapping. Genetics. 1995;141:1199–1207. - PMC - PubMed
    1. Sheng J, Jagodic M, Dahlman I, Becanovic K, Nohra R, et al. Eae19, a new locus on rat chromosome 15 regulating experimental autoimmune encephalomyelitis. Genetics. 2005 - PMC - PubMed
    1. Jagodic M, Becanovic K, Sheng JR, Wu X, Backdahl L, et al. An advanced intercross line resolves Eae18 into two narrow quantitative trait loci syntenic to multiple sclerosis candidate loci. J Immunol. 2004;173:1366–1373. - PubMed

Publication types

Feedback