Combined use of two supervised learning algorithms to model sea turtle behaviours from tri-axial acceleration data.

Accelerometers are becoming ever more important sensors in animal-attached technology, providing data that allow determination of body posture and movement and thereby helping to elucidate behaviour in animals that are difficult to observe. We sought to validate the identification of sea turtle behaviours from accelerometer signals by deploying tags on the carapace of a juvenile loggerhead (Caretta caretta), an adult hawksbill (Eretmochelys imbricata) and an adult green turtle (Chelonia mydas) at Aquarium La Rochelle, France. We recorded tri-axial acceleration at 50 Hz for each species for a full day while two fixed cameras recorded their behaviours. We identified behaviours from the acceleration data using two different supervised learning algorithms, Random Forest and Classification And Regression Tree (CART), treating the data from the adult animals as separate from the juvenile data. We achieved a global accuracy of 81.30% for the adult hawksbill and green turtle CART model and 71.63% for the juvenile loggerhead, identifying 10 and 12 different behaviours, respectively. Equivalent figures were 86.96% for the adult hawksbill and green turtle Random Forest model and 79.49% for the juvenile loggerhead, for the same behaviours. The use of Random Forest combined with CART algorithms allowed us to understand the decision rules implicated in behaviour discrimination, and thus remove or group together some 'confused' or under--represented behaviours in order to get the most accurate models. This study is the first to validate accelerometer data to identify turtle behaviours and the approach can now be tested on other captive sea turtle species.

The specific thromboxane receptor antagonist S18886: pharmacokinetic and pharmacodynamic studies.

OBJECTIVES AND PATIENTS: We conducted a multicenter double-blind pharmacokinetic/pharmacodynamic (PK/PD) study of the new oral thromboxane receptor antagonist S18886 in 30 patients with peripheral artery disease, who were randomized to receive five different oral dosages of S18886 (1, 2.5, 5, 10 or 30 mg) for 12 weeks (83 days). Primary objective was to determine the effect of S18886 on platelet aggregation ex vivo. RESULTS: Pharmacokinetics of S18886 was linear, with peak plasma levels being reached between 30 min and 2 h and a terminal half-life of 5.8-10 h. No significant accumulation of S18886 in plasma was observed after repeated dosing. The relationship between the S18886 concentration and platelet inhibition was examined in terms of U46619-induced platelet aggregation. Over the range of doses studied, there was a predictable relation between the plasma drug concentration and the degree of platelet inhibition at each dose. Maximal inhibition of U46619-induced platelet aggregation was achieved within 1 h with all oral doses of S18886, and this effect was maintained for at least 12 h. The PK/PD relationship was direct, and U46619-induced platelet aggregation was strongly inhibited by S18886 plasma concentrations above 10 ng mL(-1). This concentration was thus the minimal effective antiplatelet level in this population, and was maintained only by the dosages of 10 and 30 mg. The safety profile of S18886 was excellent, whatever the unit dose, with no attributable adverse events. CONCLUSION: The results of this study, which included modeling and simulation, help identify the minimal effective plasma concentration of S18886 required for potent antiplatelet efficacy in patients with stable peripheral arterial disease.

Study of V(1)-vascular vasopressin receptor gene microsatellite polymorphisms in human essential hypertension.

Vasopressin (AVP) actions on vascular tone and blood pressure are mainly mediated by the V(1)-vascular receptor (V(1)R). We recently reported the structure and functional expression of the human V(1)R cDNA and described the genomic characteristics, tissue expression, chromosomal localization, and regional mapping of the human V(1)R gene, AVPR1A. To test whether the V(1)R is a marker for human essential hypertension, we sequenced the human AVPR1A gene and its 5; upstream region and found several DNA microsatellite motifs. One (GT)(14)-(GA)(13)-(A)(8)microsatellite is located 2983 bp downstream of the transcription start site, within a 2.2 kbp intron interrupting the coding sequence of the receptor. Three other microsatellites are present in the 5; flanking DNA of the AVPR1A gene: a (GT)(25)dinucleotide repeat, a complex (CT)(4)-TT-(CT)(8)-(GT)(24)motif and a (GATA)(14)tetranucleotide repeat located respectively 3956 bp, 3625 bp and 553 bp upstream of the transcription start site. Analysis of these polymorphisms in 79 hypertensive and 86 normotensive subjects for the (GT)(14)-(GA)(13)-(A)(8)and the (GT)(25)motifs revealed a high percentage of heterozygosity but no difference in alleles frequencies between the two groups. A linkage study using the affected sib pair method and the (GT)(25)repeat in 446 hypertensive sib pairs from 282 French Caucasian pedigrees showed no excess of alleles sharing at the AVPR1A locus. No linkage was found in the subgroups of patients with early onset hypertension (diagnosis before age 40) or severe hypertension (diastolic blood pressure >/=100 mmHg or requirement for >/=two medications). These findings suggest that molecular variants of the V(1)R gene are not involved in unselected forms of essential hypertension.

Foods as risk factors for colorectal cancer: a case-control study in Burgundy (France).

Although the high meat-low vegetable diet is considered the reference high-risk diet for colorectal cancer, particularly in USA communities, other at-risk dietary patterns, such as high intakes of processed meat and refined carbohydrates are emerging. Little is known about risk factors for colorectal cancer in France, a country at high risk of rectal cancer and moderately high risk of colon cancer. We compared diet of colorectal cancer cases (n = 171) and general population controls (n = 309) in Burgundy (France). Categories of intake were established by sex and based on the distributions of food intakes in controls. Odds ratios for the fourth vs first quartile of intake (OR4) were 2.0 (1.1-3.6) for refined cereal products (rice, pasta and pastry), 2.4 (1.3-4.5) for delicatessen, 2.3 (1.2-4.2) for patés, 1.7 (1.1-2.8) for offal and 2.1 (1.1-4.0) for butter, lard and cream. There was no association with consumption of fresh meat (OR4 = 1.2), fish (OR4 = 1.5), egg (OR4 = 1.1) or dairy products (OR4 = 1.0). A protective effect of vegetables was only observed for left colon cancer (OR3 = 0.3; 0.1-0.6). In men, the most significant risk factors were refined cereal products, seasoning animal fats, chocolate and coffee, whereas risk factors were delicatessen, fat meat, pasta, rice, and chocolate in women. The strong association with refined cereal products is consistent with the hypothesis of a role of hyperinsulinism in colorectal carcinogenesis. The association with processed but not fresh meat suggests the importance of exogenous carcinogenesis in that area.

Structural analysis and evaluation of the 11beta-hydroxysteroid dehydrogenase type 2 (11beta-HSD2) gene in human essential hypertension.

AIM: Mutations of the 11beta-hydroxysteroid dehydrogenase type 2 (11beta-HSD2) gene cause the syndrome of apparent mineralocorticoid excess, a rare autosomal recessive form of hypertension. We therefore investigated the question of whether variants of the 11beta-HSD2 gene can contribute to genetic susceptibility to essential hypertension. SUBJECTS AND METHODS: We performed a linkage study in 162 French hypertensive sibships using the affected sib-pair method on 347 sibling pairs and a polymorphic microsatellite marker that we identified in a 30 kb cosmid clone containing the 11beta-HSD2 gene. The coding sequence, introns 2-4 and 350 bp of the 5'-flanking region of the 11beta-HSD2 gene were screened for polymorphisms by polymerase chain reaction/single-strand conformation polymorphism, and a single polymorphism, Glu178/Glu (G534A), was identified in exon 3, which did not change the encoded amino acid sequence. A case-control study was conducted on 370 hypertensive subjects with a positive family history of hypertension and 783 French subjects with hypertension with or without a family history of hypertension, compared with 313 normotensive control subjects, all of whom were analyzed for the newly identified bi-allelic polymorphism. RESULTS: Statistical analyses using the affected sib-pair method did not show significant linkage between the 11beta-HSD2 microsatellite marker and hypertension. Furthermore, no positive association with hypertension was found with the Glu178/Glu (G534A) polymorphism. CONCLUSION: Our data do not suggest that variants of the 11beta-HSD2 gene contribute substantially to essential hypertension in Caucasians.

Structural analysis and evaluation of the aldosterone synthase gene in hypertension.

Anomalies in either of the tightly linked genes encoding the enzymes CYP11B1 (11beta-hydroxylase) or CYP11B2 (aldosterone synthase) can lead to important changes in arterial pressure and are responsible for several monogenically inherited forms of hypertension. Mutations in these genes or their regulatory regions could thus contribute to genetic variation in susceptibility to essential hypertension. To test this hypothesis, we performed 2 complementary studies of the CYP11B1/CYP11B2 locus in essential hypertension. After characterizing a DNA contig containing the CYP11B1 gene and mapping the gene in the Centre d'Etudes du Polymorphisme Humain reference panel of families, we performed a linkage study with 292 hypertensive sibling pairs and a highly informative microsatellite marker near CYP11B1. We also analyzed the association of 2 frequent biallelic polymorphisms of the CYP11B2 gene, 1 in the promoter at position -344 (-344C/T) and the other, a common gene conversion in intron 2, with hypertension in 380 hypertensive patients and 293 normotensive individuals. Statistical analyses did not show significant linkage of the CYP11B1 microsatellite marker to hypertension. No positive association with hypertension was found with the gene conversion in intron 2, but a positive association with hypertension was found with the -344T allele. The hypertensive and normotensive samples differed significantly in both genotype (P=0.023) and allele frequencies (P=0.010). Our data suggest a modest contribution of the CYP11B2 gene to essential hypertension.

Evaluation of the angiotensinogen locus in human essential hypertension: a European study.

Different family and case-control studies support genetic linkage and association at the human angiotensinogen (AGT) locus with essential hypertension. To extend these previous observations, a European collaborative study of nine centers was set up to create a large resource of affected sibling pairs. The AGT locus was studied using a highly polymorphic dinucleotide repeat in the 3'-flanking region of the gene in 350 European families, comprising 630 affected sibling pairs. Statistical analyses using two different methods did not show any evidence for linkage either in the whole panel or in family subsets selected for severity or early onset of disease. Although several arguments from association studies suggest a role of the AGT gene in essential hypertension, this large family study did not replicate the initial linkage reported in smaller studies. Our results highlight the difficulty of identifying susceptibility genes by linkage analysis in complex diseases.

Genetic studies of the renin-angiotensin system in arterial hypertension associated with non-insulin-dependent diabetes mellitus.

OBJECTIVE: To determine whether angiotensinogen (AGT) and angiotensin II type 1 (AT1) receptor genes contribute to the development of arterial hypertension in members of French Caucasian families and in subjects with hypertension associated with non-insulin-dependent diabetes mellitus (NIDDM). METHODS: Sibpair linkage analyses were performed with microsatellites near the AGT and AT1 receptor genes in 179 hypertensive sibpairs from 69 NIDDM kindreds. In addition, population/association studies were performed with the M235T and T174M polymorphisms of the AGT gene, and the A1166C polymorphism of the AT1 receptor gene. RESULTS: No evidence for linkage between the AGT and AT1 receptor loci and hypertension was observed. In addition, the distributions of genotypes of AGT and AT1 receptor gene polymorphisms did not differ significantly among a group of unrelated individuals with both hypertension and NIDDM (n = 188) and three groups of unrelated control subjects with NIDDM (n = 117), hypertension (n = 75) or none of these conditions (n = 125). CONCLUSIONS: These results suggest that the AGT and AT1 receptor genes are not major genetic determinants of hypertension associated with NIDDM in this population, although we can not exclude the possibility that these loci make a minor contribution in a polygenic context.

Microsatellites and PCR primers for genetic studies and genomic sequencing of the human TSH receptor gene.

The thyroid stimulating hormone receptor (TSHR) is the main autoantigen in Graves' disease. Mutations of the TSH receptor have been implicated in various thyroid diseases. In this study, we describe three polymorphic markers localised within introns 2 and 7 of the TSH receptor gene. These markers are useful for the study of genetic diseases involving the TSH receptor. They allowed us to map precisely the TSH receptor gene on chromosome 14q31 between D14S287 and D14S68. We also describe selected primers and experimental conditions for the amplification and direct genomic sequencing of the 10 exons of the receptor gene.

[Prognostic factors in colorectal adenocarcinoma of Dukes stage B. Study of a series of population]. / Facteurs pronostiques des adénocarcinomes colorectaux de stade B de Dukes. Etude d'une série de population.

A study of prognostic factors in Dukes B colorectal adenocarcinoma was performed on a population basis in order to determine subgroups with poor prognosis which could benefit from adjuvant therapy after surgery. The study considered the 746 cases of Dukes B colorectal carcinoma diagnosed during the 1976-1988 period among Côte-d'Or residents (Burgundy, France). The overall 5-year crude survival rate was 55.7%, the corresponding relative survival rate being 73%. In the final Cox model, age, tumour extension to adjacent organs, number of examined lymph nodes and tumour size were significant prognostic factors. The corresponding multivariate relative survival model considered only tumour extension and number of examined lymph nodes as having a prognostic value. The relative risk of death was 2.3 (range: 1.5-3.3) in case of a tumor extension and 2.5 (range: 1.5-4.4) when no lymph node was found compared to surgical samples with at least 6 lymph nodes. These data should be taken into account when conceiving or analyzing future therapeutic trials.