Unsupervised Learning of Cone Spectral Classes from Natural Images

The first step in the evolution of primate trichromatic color vision was the expression of a third cone class not present in ancestral mammals. This observation motivates a fundamental question about the evolution of any sensory system: how is it possible to detect and exploit the presence of a novel sensory class? We explore this question in the context of primate color vision. We present an unsupervised learning algorithm capable of both detecting the number of spectral cone classes in a retinal mosaic and learning the class of each cone using the inter-cone correlations obtained in response to natural image input. The algorithm''s ability to classify cones is in broad agreement with experimental evidence about functional color vision for a wide range of mosaic parameters, including those characterizing dichromacy, typical trichromacy, anomalous trichromacy, and possible tetrachromacy.     

The control of copy number of IS6110 in Mycobacterium tuberculosis

Insertion sequence (IS) elements are bacterial genes that are able to transpose to different locations in the genome. These elements are often used in molecular epidemiology as genetic markers that track the spread of pathogens. Transposable elements have frequently been described as "selfish DNA" because they facilitate their own transposition, causing damage when they insert into coding regions, while contributing little if anything to the bacterial host. According to this hypothesis, the expansion of copy number of insertion sequences is opposed by negative selection against high copy numbers. From an alternative point of view, we might expect IS elements to intrinsically regulate transposition within cells, thereby limiting damage to their bacterial host. Here, we report evidence that the copy number of IS6110 in Mycobacterium tuberculosis is controlled by selection against the element. We first construct 12 different models of marker change resulting from a combination of possible transposition functions and selective regimes. We then compute the Akaike Information Criterion for each model to identify the models that best explain data consisting of serial isolates of M. tuberculosis genotyped with IS6110. We find that the best performing models all include selection against the accumulation of copies. Specifically, our analysis points to the interaction of separate copies of the element causing lethal effects. We discuss the implications of these findings for genome evolution and molecular epidemiology.     

A format for phylogenetic placements

We have developed a unified format for phylogenetic placements, that is, mappings of environmental sequence data (e.g., short reads) into a phylogenetic tree. We are motivated to do so by the growing number of tools for computing and post-processing phylogenetic placements, and the lack of an established standard for storing them. The format is lightweight, versatile, extensible, and is based on the JSON format, which can be parsed by most modern programming languages. Our format is already implemented in several tools for computing and post-processing parsimony- and likelihood-based phylogenetic placements and has worked well in practice. We believe that establishing a standard format for analyzing read placements at this early stage will lead to a more efficient development of powerful and portable post-analysis tools for the growing applications of phylogenetic placement.     

The Relationship between Imputation Error and Statistical Power in Genetic Association Studies in Diverse Populations

Genotype-imputation methods provide an essential technique for high-resolution genome-wide association (GWA) studies with millions of single-nucleotide polymorphisms. For optimal design and interpretation of imputation-based GWA studies, it is important to understand the connection between imputation error and power to detect associations at imputed markers. Here, using a 2 × 3 chi-square test, we describe a relationship between genotype-imputation error rates and the sample-size inflation required for achieving statistical power at an imputed marker equal to that obtained if genotypes at the marker were known with certainty. Surprisingly, typical imputation error rates (∼2%–6%) lead to a large increase in the required sample size (∼10%–60%), and in some African populations whose genotypes are particularly difficult to impute, the required sample-size increase is as high as ∼30%–150%. In most populations, each 1% increase in imputation error leads to an increase of ∼5%–13% in the sample size required for maintaining power. These results imply that in GWA sample-size calculations investigators will need to account for a potentially considerable loss of power from even low levels of imputation error and that development of additional genomic resources that decrease imputation error will translate into substantial reduction in the sample sizes needed for imputation-based detection of the variants that underlie complex human diseases.     

PKCθ is required for hemostasis and positive regulation of thrombin-induced platelet aggregation and α-granule secretion

Platelet activation due to vascular injury is essential for hemostatic plug formation, and is mediated by agonists, such as thrombin, which trigger distinct receptor-coupled signaling pathways. Thrombin is a coagulation protease, which activates G protein-coupled protease-activated receptors (PARs) on the surface of platelets. We found that C57BL/6J and BALB/C mice that are deficient in protein kinase C θ (PKCθ), exhibit an impaired hemostasis, and prolonged bleeding following vascular injury. In addition, murine platelets deficient in PKCθ displayed an impaired thrombin-induced platelet activation and aggregation response. Lack of PKCθ also resulted in impaired α-granule secretion, as demonstrated by the low surface expression of CD62P, in thrombin-stimulated platelets. Since PAR4 is the only mouse PAR receptor that delivers thrombin-induced activation signals in platelets, our results suggest that PKCθ is a critical effector molecule in the PAR4-linked signaling pathways and in the regulation of normal hemostasis in mice.     

Terpenoids from Zingiber officinale (Ginger) Induce Apoptosis in Endometrial Cancer Cells through the Activation of p53

Novel strategies are necessary to improve chemotherapy response in advanced and recurrent endometrial cancer. Here, we demonstrate that terpenoids present in the Steam Distilled Extract of Ginger (SDGE) are potent inhibitors of proliferation of endometrial cancer cells. SDGE, isolated from six different batches of ginger rhizomes, consistently inhibited proliferation of the endometrial cancer cell lines Ishikawa and ECC-1 at IC₅₀ of 1.25 µg/ml. SDGE also enhanced the anti-proliferative effect of radiation and cisplatin. Decreased proliferation of Ishikawa and ECC-1 cells was a direct result of SDGE-induced apoptosis as demonstrated by FITC-Annexin V staining and expression of cleaved caspase 3. GC/MS analysis identified a total of 22 different terpenoid compounds in SDGE, with the isomers neral and geranial constituting 30–40%. Citral, a mixture of neral and geranial inhibited the proliferation of Ishikawa and ECC-1 cells at an IC₅₀ 10 µM (2.3 µg/ml). Phenolic compounds such as gingerol and shogaol were not detected in SDGE and 6-gingerol was a weaker inhibitor of the proliferation of the endometrial cancer cells. SDGE was more effective in inducing cancer cell death than citral, suggesting that other terpenes present in SDGE were also contributing to endometrial cancer cell death. SDGE treatment resulted in a rapid and strong increase in intracellular calcium and a 20–40% decrease in the mitochondrial membrane potential. Ser-15 of p53 was phosphorylated after 15 min treatment of the cancer cells with SDGE. This increase in p53 was associated with 90% decrease in Bcl2 whereas no effect was observed on Bax. Inhibitor of p53, pifithrin-α, attenuated the anti-cancer effects of SDGE and apoptosis was also not observed in the p53^neg SKOV-3 cells. Our studies demonstrate that terpenoids from SDGE mediate apoptosis by activating p53 and should be therefore be investigated as agents for the treatment of endometrial cancer.     

Across bacterial phyla, distantly-related genomes with similar genomic GC content have similar patterns of amino acid usage

The GC content of bacterial genomes ranges from 16% to 75% and wide ranges of genomic GC content are observed within many bacterial phyla, including both gram negative and gram positive phyla. Thus, divergent genomic GC content has evolved repeatedly in widely separated bacterial taxa. Since genomic GC content influences codon usage, we examined codon usage patterns and predicted protein amino acid content as a function of genomic GC content within eight different phyla or classes of bacteria. We found that similar patterns of codon usage and protein amino acid content have evolved independently in all eight groups of bacteria. For example, in each group, use of amino acids encoded by GC-rich codons increased by approximately 1% for each 10% increase in genomic GC content, while the use of amino acids encoded by AT-rich codons decreased by a similar amount. This consistency within every phylum and class studied led us to conclude that GC content appears to be the primary determinant of the codon and amino acid usage patterns observed in bacterial genomes. These results also indicate that selection for translational efficiency of highly expressed genes is constrained by the genomic parameters associated with the GC content of the host genome.     

Germ cell apoptosis in the testes of normal stallions

Apoptosis in testicular germ cells has been demonstrated in many mammalian species. However, little is known about the stallion (Equus caballus) and rates of apoptosis during spermatogenesis. Morphological and biochemical features of apoptosis reported in other species were used to confirm that the TdT-mediated dUTP Nick end labeling (TUNEL) assay is an acceptable method for identification and quantification of apoptotic germ cells in histological tissue sections from stallion testis. Seminiferous tubules from eight stallions with normal testis size and semen quality were evaluated according to stage of seminiferous epithelium to determine the germ cell types and stages where apoptosis most commonly occurs. Spermatogonia and spermatocytes were the most common germ cell types labeled by the TUNEL assay. A low rate of round and elongated spermatids were labeled by the TUNEL assay. Mean numbers of TUNEL-positive germ cells per 100 Sertoli cell nuclei were highest in stages IV (15.5 +/- 1.0) and V (13.5 +/- 1.1) of the seminiferous epithelial cycle (P < 0.001). An intermediate level of apoptosis was detected in stage VI (P < 0.02). These stages (IV-VI) correspond to meiotic divisions of primary spermatocytes and mitotic proliferation of B1 and B2 spermatogonia. Establishing basal levels of germ cell apoptosis is a critical step towards understanding fertility and the role of apoptosis in regulating germ cell numbers during spermatogenesis.