Influence of immune-relevant genotype on the reproductive success of a salmonid alternative mating strategy.

Major histocompatibility complex (MHC) and immune-relevant gene markers were used to evaluate differences in reproductive success (RS) among naturally spawning coho salmon Oncorhynchus kisutch mate pairs involving an alternative male reproductive phenotype, known as jacks. These mate pairs included both hatchery-reared and wild origin fish such that three classes were evaluated in two consecutive years (2005 and 2006) using a previously constructed multigenerational genetic pedigree: wild × wild (W × W), hatchery × hatchery (H × H) and wild × hatchery (W × H). Oncorhynchus kisutch jack mate pairs mated randomly based on immune-relevant genotype in both years; a result consistent with the opportunistic mating strategy of jacks. An association between greater number of alleles shared at three immune-relevant gene markers and increased RS was found for: W × H mate pairs in 2005 (BHMS429), W × H pairs in 2006 (SsalR016TKU) and W × W pairs in 2006 (OMM3085). No correlation between immune gene diversity and RS was found for H × H pairs in either year. The results suggest that the influence of immune-relevant genotype on mating success may be different for jacks when compared with previous studies of large adult male O. kisutch.

Probability machines: consistent probability estimation using nonparametric learning machines.

BACKGROUND: Most machine learning approaches only provide a classification for binary responses. However, probabilities are required for risk estimation using individual patient characteristics. It has been shown recently that every statistical learning machine known to be consistent for a nonparametric regression problem is a probability machine that is provably consistent for this estimation problem. OBJECTIVES: The aim of this paper is to show how random forests and nearest neighbors can be used for consistent estimation of individual probabilities. METHODS: Two random forest algorithms and two nearest neighbor algorithms are described in detail for estimation of individual probabilities. We discuss the consistency of random forests, nearest neighbors and other learning machines in detail. We conduct a simulation study to illustrate the validity of the methods. We exemplify the algorithms by analyzing two well-known data sets on the diagnosis of appendicitis and the diagnosis of diabetes in Pima Indians. RESULTS: Simulations demonstrate the validity of the method. With the real data application, we show the accuracy and practicality of this approach. We provide sample code from R packages in which the probability estimation is already available. This means that all calculations can be performed using existing software. CONCLUSIONS: Random forest algorithms as well as nearest neighbor approaches are valid machine learning methods for estimating individual probabilities for binary responses. Freely available implementations are available in R and may be used for applications.

Duplicated Clock genes with unique polyglutamine domains provide evidence for nonhomologous recombination in Chinook salmon (Oncorhynchus tshawytscha).

Circadian rhythms underlie diverse life functions ranging from cellular activities to behavior. Multiple clock genes play a central role in the generation of these rhythms. We partially characterized two copies of the Clock gene from Chinook salmon (Oncorhynchus tshawytscha), OtsClock1a and OtsClock1b. The 6,460 bp OtsClock1a sequence contains 16 exons, 15 introns and encompasses three highly conserved domains indicating it is a novel member of the bHLH-PAS superfamily of transcription factors. The second copy, OtsClock1b, consists of five exons and five introns spanning 1,945 bp. A polyglutamine repeat motif (PolyQ), characteristic of a majority of CLOCK proteins, is present in both OTSCLOCK1a and OTSCLOCK1b. However, the Chinook PolyQ domains are uniquely positioned inside the gene. Interestingly, a 1,200 bp non-coding segment located downstream of the OtsClock1a PolyQ domain is absent from OtsClock1b. This insertion/deletion is 91% similar to the Salmo salar Transferrin gene. A phylogenetic analysis of 11 CLOCK proteins shows that OtsClock1a and OtsClock1b are paralogs which likely arose subsequent to the salmonid genome-wide duplication event. Ultimately, the Chinook salmon Clock genes are key components to our understanding the genetic mechanisms underlying temporally regulated life history traits in Pacific salmonids.

Quantitative trait loci for spawning date and body weight in rainbow trout: testing for conserved effects across ancestrally duplicated chromosomes.

We incorporated 69 microsatellite loci into an existing data set of 132 markers to test for quantitative trait loci (QTLs) affecting spawning date and body weight in a backcross between two outbred strains of rainbow trout (Oncorhynchus mykiss). Twenty-six linkage groups were identified and synteny of duplicated microsatellite markers was used to confirm 13 homeologous chromosome pairs. Gene-centromere data were used to localize the centromeres for 13 linkage groups whose orientations were previously unknown. We applied a combination of interval mapping and single marker analysis to the segregating maternal and paternal alleles at 201 microsatellite loci. Four spawning date QTLs with suggestive evidence for an additional two QTLs were detected in female trout spawning at 3 and 4 years of age. Similarly we detected three QTLs for body weight in females at 2 years of age plus four suggestive QTLs for this trait. We found marginal evidence that three pairs of ancestral homeologues contained detectable QTLs for the same trait. In one of the three pairs of homeologues, the duplicated QTL regions mapped to the same relative chromosomal location, while the exact localization of the QTL position in one of the other pairs was difficult to infer since it was based on data from a male-derived map. The existing data were unable to refute a hypothesis that duplicated functional genes will be maintained within the telomeric regions of salmonids due to preferential male-mediated crossing over in this region. Two of the four spawning date QTLs were detected on linkage groups with unknown homeologous relationships. QTLs with possible pleiotropic effects on both spawning date and body size were localized to two linkage groups.

Variation of hepatic fibrosis and granuloma size among mouse strains infected with Schistosoma mansoni.

To investigate the relation between the size of circumoval granulomas and hepatic fibrosis, a variety of mouse strains infected with Schistosoma mansoni were examined and the number of eggs in the tissues, the fibrotic responses to the eggs, and the volume of the granulomas were determined. Marked differences in granuloma volume and in hepatic fibrosis were found between mouse strains, and those strains with the largest granulomas also showed the most hepatic fibrosis. On the other hand no significant correlation between granuloma size and hepatic fibrosis was found in the progeny of the F2 generation and backcrosses between F1 mice and the parental strains when crosses were made between Nmri mice (high granuloma volume and high fibrosis) and C57BL/6 mice (low granuloma volume and low fibrosis). Hepatic fibrosis per egg decreased with increasing infection intensity while granuloma volume was unaffected, indicating that fibrosis and granuloma size are at least in part modulated by different factors. The number of eggs found in the tissues per worm pair and the proportion of eggs in the liver also decreased as infection intensity increased. Some influence of the major histocompatibility complex on both granuloma size and fibrosis was found. Congenic mice on the C57BL/10 and C3H/HeSn backgrounds showed larger granulomas in H-2b than in H-2k mice, but no such correlation was found in comparing C57BL/6 mice with B6.H-2k mice. Less hepatic fibrosis was found in B10.M (H-2f), B10.SM (H-2v), and B10.RIII (H-2r) animals than in C57BL/10 mice. The regulation of granuloma size and of hepatic fibrosis is clearly complex and involves genes both outside of and within the major histocompatibility complex.