Is Demography Destiny? Application of Machine Learning Techniques to Accurately Predict Population Health Outcomes from a Minimal Demographic Dataset

For years, we have relied on population surveys to keep track of regional public health statistics, including the prevalence of non-communicable diseases. Because of the cost and limitations of such surveys, we often do not have the up-to-date data on health outcomes of a region. In this paper, we examined the feasibility of inferring regional health outcomes from socio-demographic data that are widely available and timely updated through national censuses and community surveys. Using data for 50 American states (excluding Washington DC) from 2007 to 2012, we constructed a machine-learning model to predict the prevalence of six non-communicable disease (NCD) outcomes (four NCDs and two major clinical risk factors), based on population socio-demographic characteristics from the American Community Survey. We found that regional prevalence estimates for non-communicable diseases can be reasonably predicted. The predictions were highly correlated with the observed data, in both the states included in the derivation model (median correlation 0.88) and those excluded from the development for use as a completely separated validation sample (median correlation 0.85), demonstrating that the model had sufficient external validity to make good predictions, based on demographics alone, for areas not included in the model development. This highlights both the utility of this sophisticated approach to model development, and the vital importance of simple socio-demographic characteristics as both indicators and determinants of chronic disease.     

Collective cell migration of smooth muscle and endothelial cells: impact of injury versus non-injury stimuli

Background

Cell migration is a vital process for growth and repair. In vitro migration assays, utilized to study cell migration, often rely on physical scraping of a cell monolayer to induce cell migration. The physical act of scrape injury results in numerous factors stimulating cell migration – some injury-related, some solely due to gap creation and loss of contact inhibition. Eliminating the effects of cell injury would be useful to examine the relative contribution of injury versus other mechanisms to cell migration. Cell exclusion assays can tease out the effects of injury and have become a new avenue for migration studies. Here, we developed two simple non-injury techniques for cell exclusion: 1) a Pyrex® cylinder - for outward migration of cells and 2) a polydimethylsiloxane (PDMS) insert - for inward migration of cells. Utilizing these assays smooth muscle cells (SMCs) and human umbilical vein endothelial cells (HUVECs) migratory behavior was studied on both polystyrene and gelatin-coated surfaces.

Results

Differences in migratory behavior could be detected for both smooth muscle cells (SMCs) and endothelial cells (ECs) when utilizing injury versus non-injury assays. SMCs migrated faster than HUVECs when stimulated by injury in the scrape wound assay, with rates of 1.26 % per hour and 1.59 % per hour on polystyrene and gelatin surfaces, respectively. The fastest overall migration took place with HUVECs on a gelatin-coated surface, with the in-growth assay, at a rate of 2.05 % per hour. The slowest migration occurred with the same conditions but on a polystyrene surface at a rate of 0.33 % per hour.

Conclusion

For SMCs, injury is a dominating factor in migration when compared to the two cell exclusion assays, regardless of the surface tested: polystyrene or gelatin. In contrast, the migrating surface, namely gelatin, was a dominating factor for HUVEC migration, providing an increase in cell migration over the polystyrene surface. Overall, the cell exclusion assays - the in-growth and out-growth assays, provide a means to determine pure migratory behavior of cells in comparison to migration confounded by cell wounding and injury.

     

Large-Scale Analysis of Kinase Signaling in Yeast Pseudohyphal Development Identifies Regulation of Ribonucleoprotein Granules

Yeast pseudohyphal filamentation is a stress-responsive growth transition relevant to processes required for virulence in pathogenic fungi. Pseudohyphal growth is controlled through a regulatory network encompassing conserved MAPK (Ste20p, Ste11p, Ste7p, Kss1p, and Fus3p), protein kinase A (Tpk2p), Elm1p, and Snf1p kinase pathways; however, the scope of these pathways is not fully understood. Here, we implemented quantitative phosphoproteomics to identify each of these signaling networks, generating a kinase-dead mutant in filamentous S. cerevisiae and surveying for differential phosphorylation. By this approach, we identified 439 phosphoproteins dependent upon pseudohyphal growth kinases. We report novel phosphorylation sites in 543 peptides, including phosphorylated residues in Ras2p and Flo8p required for wild-type filamentous growth. Phosphoproteins in these kinase signaling networks were enriched for ribonucleoprotein (RNP) granule components, and we observe co-localization of Kss1p, Fus3p, Ste20p, and Tpk2p with the RNP component Igo1p. These kinases localize in puncta with GFP-visualized mRNA, and KSS1 is required for wild-type levels of mRNA localization in RNPs. Kss1p pathway activity is reduced in lsm1Δ/Δ and pat1Δ/Δ strains, and these genes encoding P-body proteins are epistatic to STE7. The P-body protein Dhh1p is also required for hyphal development in Candida albicans. Collectively, this study presents a wealth of data identifying the yeast phosphoproteome in pseudohyphal growth and regulatory interrelationships between pseudohyphal growth kinases and RNPs.     

The Shelterin TIN2 Subunit Mediates Recruitment of Telomerase to Telomeres

Dyskeratosis Congenita (DC) is a heritable multi-system disorder caused by abnormally short telomeres. Clinically diagnosed by the mucocutaneous symptoms, DC patients are at high risk for bone marrow failure, pulmonary fibrosis, and multiple types of cancers. We have recapitulated the most common DC-causing mutation in the shelterin component TIN2 by introducing a TIN2-R282H mutation into cultured telomerase-positive human cells via a knock-in approach. The resulting heterozygous TIN2-R282H mutation does not perturb occupancy of other shelterin components on telomeres, result in activation of telomeric DNA damage signaling or exhibit other characteristics indicative of a telomere deprotection defect. Using a novel assay that monitors the frequency and extension rate of telomerase activity at individual telomeres, we show instead that telomerase elongates telomeres at a reduced frequency in TIN2-R282H heterozygous cells; this recruitment defect is further corroborated by examining the effect of this mutation on telomerase-telomere co-localization. These observations suggest a direct role for TIN2 in mediating telomere length through telomerase, separable from its role in telomere protection.     

Contribution of the Two Genes Encoding Histone Variant H3.3 to Viability and Fertility in Mice

Histones package DNA and regulate epigenetic states. For the latter, probably the most important histone is H3. Mammals have three near-identical H3 isoforms: canonical H3.1 and H3.2, and the replication-independent variant H3.3. This variant can accumulate in slowly dividing somatic cells, replacing canonical H3. Some replication-independent histones, through their ability to incorporate outside S-phase, are functionally important in the very slowly dividing mammalian germ line. Much remains to be learned of H3.3 functions in germ cell development.Histone H3.3 presents a unique genetic paradigm in that two conventional intron-containing genes encode the identical protein. Here, we present a comprehensive analysis of the developmental effects of null mutations in each of these genes. H3f3a mutants were viable to adulthood. Females were fertile, while males were subfertile with dysmorphic spermatozoa. H3f3b mutants were growth-deficient, dying at birth. H3f3b heterozygotes were also growth-deficient, with males being sterile because of arrest of round spermatids. This sterility was not accompanied by abnormalities in sex chromosome inactivation in meiosis I. Conditional ablation of H3f3b at the beginning of folliculogenesis resulted in zygote cleavage failure, establishing H3f3b as a maternal-effect gene, and revealing a requirement for H3.3 in the first mitosis. Simultaneous ablation of H3f3a and H3f3b in folliculogenesis resulted in early primary oocyte death, demonstrating a crucial role for H3.3 in oogenesis.These findings reveal a heavy reliance on H3.3 for growth, gametogenesis, and fertilization, identifying developmental processes that are particularly susceptible to H3.3 deficiency. They also reveal partial redundancy in function of H3f3a and H3f3b, with the latter gene being generally the most important.     

Effects of Initial Values and Convergence Criterion in the Two-Parameter Logistic Model When Estimating the Latent Distribution in BILOG-MG 3

Parameters of the two-parameter logistic model are generally estimated via the expectation-maximization algorithm, which improves initial values for all parameters iteratively until convergence is reached. Effects of initial values are rarely discussed in item response theory (IRT), but initial values were recently found to affect item parameters when estimating the latent distribution with full non-parametric maximum likelihood. However, this method is rarely used in practice. Hence, the present study investigated effects of initial values on item parameter bias and on recovery of item characteristic curves in BILOG-MG 3, a widely used IRT software package. Results showed notable effects of initial values on item parameters. For tighter convergence criteria, effects of initial values decreased, but item parameter bias increased, and the recovery of the latent distribution worsened. For practical application, it is advised to use the BILOG default convergence criterion with appropriate initial values when estimating the latent distribution from data.