The branch-site test of positive selection is surprisingly robust but lacks power under synonymous substitution saturation and variation in GC

Positive selection is widely estimated from protein coding sequence alignments by the nonsynonymous-to-synonymous ratio omega. Increasingly elaborate codon models are used in a likelihood framework for this estimation. Although there is widespread concern about the robustness of the estimation of the omega ratio, more efforts are needed to estimate this robustness, especially in the context of complex models. Here, we focused on the branch-site codon model. We investigated its robustness on a large set of simulated data. First, we investigated the impact of sequence divergence. We found evidence of underestimation of the synonymous substitution rate for values as small as 0.5, with a slight increase in false positives for the branch-site test. When dS increases further, underestimation of dS is worse, but false positives decrease. Interestingly, the detection of true positives follows a similar distribution, with a maximum for intermediary values of dS. Thus, high dS is more of a concern for a loss of power (false negatives) than for false positives of the test. Second, we investigated the impact of GC content. We showed that there is no significant difference of false positives between high GC (up to similar to 80%) and low GC (similar to 30%) genes. Moreover, neither shifts of GC content on a specific branch nor major shifts in GC along the gene sequence generate many false positives. Our results confirm that the branch-site is a very conservative test.

Molar substitution and C2/C6 ratio of hydroxyethyl starch: influence of blood coagulation

Résumé Cette étude a démontré l'effet individuel sur la coagulation sanguine humaine des deux principales caractéristiques de la molécule d'hydroxyéthylamidon (HES) : la substitution molaire et le rapport C2/C6. L'analyse par thrombélastographe (TEG®) indique que la molécule de HES dont la substitution molaire est de 0.42 et le rapport C2/C6 de 2.7 a le moins d'effet sur la coagulation sanguine chez l'être humain. Objectifs de l'étude Le développement d'hydroxyéthylamidons (HES) qui ont à la fois peu d'impact sur la coagulation sanguine et une longue persistance intravasculaire est d'un grand intérêt clinique. Une précédente étude in vitro a démontré qu'une solution de HES de haut poids moléculaire et de bas degré de substitution molaire ne compromettait pas plus la coagulation sanguine qu'une solution HES de poids moléculaire moyen (1). La présente étude examine l'effet individuel de la substitution molaire et du rapport C2/C6 d'une solution de HES de haut poids moléculaire (700 kDa) sur la coagulation sanguine. Matériel et méthode Nous avons prélevé du sang chez 30 adultes en bonne santé; le sang a été mélangé avec 6 solutions de HES qui diffèrent par leur degré de substitution molaire (0.42 et 0.51) et leur rapport C2/C6 (2.7, 7 et 14) à trois degrés de dilution : 20%, 40% et 60%. Les échantillons ont ensuite été analysés par thrombélastographe. Les données ont été étudiées par analyse de variance à trois voies pour mesures répétées sur une voie (dilution). Résultats Plus la substitution molaire est élevée, plus la coagulation sanguine est compromise et ce concernant tous les paramètres du TEG® (tous les p sont < à 0.05). La solution HES avec le rapport C2/C6 le plus bas a l'effet le moins prononcé sur le temps r (p<0.001), l'angle α (p=0.003) et l'Index de Coagulation CI (p<0.001) ; on n'a pas observé d'effet sur le temps k (p=0.513) et l'amplitude maximale (p=0.699) concernant ce paramètre. Conclusion L'analyse par thrombélastographe révèle qu'une molécule de HES avec une substitution molaire de 0.42 et un rapport C2/C6 de 2.7 a un effet minimal sur la coagulation sanguine humaine in vitro.

Potential health-care service utilization substitution effects induced by case management for emergency department frequent users

BACKGROUND: In most of the emergency departments (ED) in developed countries, a subset of patients visits the ED frequently. Despite their small numbers, these patients are the source of a disproportionally high number of all ED visits, and use a significant proportion of healthcare resources. They place a heavy economic burden on hospital and healthcare systems budgets overall. Several interventions have been carried out to improve the management of these ED frequent users. Case management has been shown in some North American studies to reduce ED utilization and costs. In these studies, cost analyses have been carried out from the hospital perspective without examining the costs induced by healthcare consumed in the community. However, case management might reduce ED visits and costs from the hospital's perspective, but induce substitution effects, and increase health service utilization outside the hospital. This study examined if an interdisciplinary case-management intervention-compared to standard ED care -reduced costs generated by frequent ED users not only from the hospital perspective, but also from the healthcare system perspective-that is, from a broader perspective taking into account the costs of healthcare services used outside the hospital. METHODS: In this randomized controlled trial, 250 adult frequent emergency department users (5 or more visits during the previous 12 months) who visited the ED of the University Hospital of Lausanne, Switzerland, between May 2012 and July 2013 were allocated to one of two groups: case management intervention (CM) or standard ED care (SC), and followed up for 12 months. Depending on the perspective of the analysis, costs were evaluated differently. For the analysis from the hospital's perspective, the true value of resources used to provide services was used as a cost estimate. These data were obtained from the hospital's analytical accounting system. For the analysis from the health-care system perspective, all health-care services consumed by users and charged were used as an estimate of costs. These data were obtained from health insurance providers for a subsample of participants. To allow comparisons in a same time period, individual monthly average costs were calculated. Multivariate linear models including a fixed effect "group" were run using socio-demographic characteristics and health-related variables as controlling variables (age, gender, educational level, citizenship, marital status, somatic and mental health problems, and risk behaviors).