Dissecting High-Dimensional Phenotypes with Bayesian Sparse Factor Analysis of Genetic Covariance Matrices

Quantitative genetic studies that model complex, multivariate phenotypes are important for both evolutionary prediction and artificial selection. For example, changes in gene expression can provide insight into developmental and physiological mechanisms that link genotype and phenotype. However, classical analytical techniques are poorly suited to quantitative genetic studies of gene expression where the number of traits assayed per individual can reach many thousand. Here, we derive a Bayesian genetic sparse factor model for estimating the genetic covariance matrix (G-matrix) of high-dimensional traits, such as gene expression, in a mixed-effects model. The key idea of our model is that we need consider only G-matrices that are biologically plausible. An organism’s entire phenotype is the result of processes that are modular and have limited complexity. This implies that the G-matrix will be highly structured. In particular, we assume that a limited number of intermediate traits (or factors, e.g., variations in development or physiology) control the variation in the high-dimensional phenotype, and that each of these intermediate traits is sparse – affecting only a few observed traits. The advantages of this approach are twofold. First, sparse factors are interpretable and provide biological insight into mechanisms underlying the genetic architecture. Second, enforcing sparsity helps prevent sampling errors from swamping out the true signal in high-dimensional data. We demonstrate the advantages of our model on simulated data and in an analysis of a published Drosophila melanogaster gene expression data set.     

Molecular adaptation of a leaf-eating bird: stomach lysozyme of the hoatzin

This report describes a lysozyme expressed at high levels in the stomach of the hoatzin, the only known foregut-fermenting bird. Evolutionary comparison places it among the calcium-binding lysozymes rather than among the conventional types. Conventional lysozymes were recruited as digestive enzymes twice in the evolution of mammalian foregut fermenters, and these independently recruited lysozymes share convergent structural changes attributed to selective pressures in the stomach. Biochemical convergence and parallel amino acid replacements are observed in the hoatzin stomach lysozyme even though it has a different genetic origin from the mammalian examples and has undergone more than 300 million years of independent evolution.      

In vivo analysis of fracture toughness of thyroid gland tumors

Background

Human solid tumors that are hard or firm on physical palpation are likely to be cancerous, a clinical maxim that has been successfully applied to cancer screening programs, such as breast self-examination. However, the biological relevance or prognostic significance of tumor hardness remains poorly understood. Here we present a fracture mechanics based in vivo approach for characterizing the fracture toughness of biological tissue of human thyroid gland tumors.

Methods

In a prospective study, 609 solid thyroid gland tumors were percutaneously probed using standard 25 gauge fine needles, their tissue toughness ranked on the basis of the nature and strength of the haptic force feedback cues, and subjected to standard fine needle biopsy. The tumors' toughness rankings and final cytological diagnoses were combined and analyzed. The interpreting cytopathologist was blinded to the tumors' toughness rankings.

Results

Our data showed that cancerous and noncancerous tumors displayed remarkable haptically distinguishable differences in their material toughness.

Conclusion

The qualitative method described here, though subject to some operator bias, identifies a previously unreported in vivo approach to classify fracture toughness of a solid tumor that can be correlated with malignancy, and paves the way for the development of a mechanical device that can accurately quantify the tissue toughness of a human tumor.     

Restriction-map variation with the yellow-achaete-scute region in five populations of Drosophila melanogaster

It has been proposed that the degree of recombination for a genomic region will affect the level of both nucleotide heterozygosity and the density of transposable elements. Both features of genomic diversity have been examined in a number of recent reports for regions undergoing relatively normal levels of recombination in Drosophila melanogaster. In this study the genomic variation associated with yellow-achaete- scute loci located at the tip of the X chromosome is examined by six- cutter restriction mapping. In this region, as usual for regions adjacent to telomeres, crossing-over is dramatically reduced, and published studies of visible mutants indicate extremely little restriction-map variation. Eight six-cutter restriction endonucleases were used to locate sequence variation in 14- and 16.5-kb regions in 109 lines sampled from North America, Africa, and Europe. The overall level of heterozygosity is estimated as 0.29%. Nine large insertions, all presumed to be transposable elements, were observed. Base-pair heterozygosity appears to be reduced compared with regions having normal levels of recombination. The estimated heterozygosity is much higher than reported in earlier studies of restriction-map variation among visible mutations in the complex. The incidence of large insertions is not elevated compared with that in other regions of the genome. This suggests that asymmetric synapsis and exchange is not an important mechanism for the elimination of transposable elements.      

Current practice in transferring critically ill patients among hospitals in the west of Scotland.

OBJECTIVE--To identify the requirements of an interhospital transfer service for critically ill patients. DESIGN--Retrospective survey of the current functions of a specialist interhospital transfer team from data collected at the time of transfer and from records of intensive care unit. SETTING--Mobile intensive care unit based at a tertiary referral centre, which serves the west of Scotland. PATIENTS--All critically ill patients (378) transferred between hospitals by the unit from 1986 to 1988. RESULTS--365 Patients were transferred by road and 13 by air. There was a wide variation in age (range 6 weeks to 87 years), diagnosis, reason for transfer, support required, and distance travelled. Most patients (232) were transferred for respiratory or cardiovascular support; 100 were trauma cases. 300 Patients (79%) were mechanically ventilated during transfer. No patient died in transit, although the eventual mortality was 28% (105 patients). Mortality was significantly higher in patients transferred from hospitals with intensive care units than from those without (38% (125 patients) v 23% (253); p less than 0.005). IMPLICATIONS--Safe interhospital transfer of critically ill patients is feasible; the high eventual mortality in some patient groups emphasises the need for accurate prediction of outcome if inappropriate transfer is to be avoided. The findings may help in organising secondary transfer services in future.