A Comparative Study of Scores for Correspondence Analysis with Ordered Categories

Ordered categorical data can be analysed using correspondence analysis with the ordered categories taken into consideration. Such an analysis was proposed by Beh (1997) and uses orthogonal polynomials which require the input of a scoring scheme to reflect the ordered structure of the categories. This method of correspondence analysis visualises the relationship between the categories, in terms of the location, dispersion and higher order components. The impact of the scoring method on the orthogonal polynomials, and hence upon the correspondence plot and other output of the analysis should therefore be considered. This paper aims at identifying this impact by considering four scoring schemes: integer valued (natural) scores, midrank scores, Nishisato scores and singular vectors from the classical correspondence analysis of the data. It is shown that while the latter two maximise the location component, generally there is little difference when comparing them with the output of the former two scoring schemes. A simple comparative study of profile co-ordinates using different scoring schemes is also discussed.     

Association of CELSR2, APOB100, ABCG5/8, LDLR,and APOE polymorphisms and their genetic risks with lipids among the Thai subjects

BackgroundHypercholesterolemia is a common cardiovascular risk factor. The aim of this study was to investigate the association of CELSR2 (rs629301), APOB100 (rs1367117), ABCG5/8 (rs6544713), LDLR (rs6511720), and APOE (rs429358, rs7412) polymorphisms, and their genetic risk scores with lipids among Thai subjects.MethodsA total of 459 study subjects (184 males, and 275 females) were enrolled. Blood pressure, serum lipids, and fasting blood sugar were measured. CELSR2 (rs629301), APOB100 (rs1367117), ABCG5/8 (rs6544713), and LDLR (rs6511720) polymorphisms were analyzed using PCR-HRM. APOE (rs429358, rs7412) polymorphism was analyzed using PCR-RFLP.ResultsTotal cholesterol (TC) levels were significantly higher in APOB100 AA genotype compared with GG, or AA + AG genotypes in total subjects. In addition, significantly higher concentrations of TC and low density lipoprotein cholesterol (LDL-C) were observed in APOE4 carriers compared to APOE2 carriers in total subjects, males, and females. The significantly higher concentrations of TC were observed in APOE4 carriers compared to APOE3 carriers in females. Moreover, the concentrations of TC, and LDL-C were significantly increased with genetic risk scores of APOB100, and APOE polymorphisms in total subjects, and females. There was no association between CELSR2 (rs629301), ABCG5/8 (rs6544713), and LDLR (rs6511720) polymorphisms and serum lipids.ConclusionAPOB100 (rs1367117), and APOE (rs429358, rs7412) but not CELSR2 (rs629301), ABCG5/8 (rs6544713), and LDLR (rs6511720) polymorphisms were associated with serum lipids. The cumulative risk alleles of APOB100 (rs1367117), and APOE (rs429358, rs7412) polymorphisms could enhance the elevated concentrations of TC, and LDL-C, and they may be used to predict severity of hypercholesterolemia among Thai subjects.     

On the Use of Factor Scores for Prediction

There are two types of indeterminacy problems in factor analysis: the indeterminacy of the parameters, which is also known as the problem of identification, and the indeterminacy of the factor scores. The second problem, which will be treated here, is of great importance to the social sciences, given that the factor analysis model was historically developed with the theory of measuring intelligence, where the main interest was to determine one person's intelligence numerically. As a consequence of the indeterminacy of factor scores we will show, by means of a numerical example, that any criterion, including the dates of easter sunday, can be perfectly predicted by the intelligence factors.     

The Determination of Sample Sizes in the Comparison of Two Multinomial Proportions from Ordered Categories

We consider sample size determination for ordered categorical data when the alternative assumption is the proportional odds model. In this paper the sample size formula proposed by Whitehead (Statistics in Medicine, 12 , 2257–2271, 1993) is compared with the methods based on exact and asymptotic linear rank tests with Wilcoxon and trend scores. We show that Whitehead's formula, which is based on a normal approximation, works well when the sample size is moderate to large but recommend the exact method with Wilcoxon scores for small sample sizes. The consequences of misspecification in models are also investigated.     

A Review and Evaluation of Plant Protection Product Ranking Tools Used in Agriculture

Pesticides used in the agricultural sector could potentially negatively impact the ecosystem, and consequently human and animal health. At the European level, legislation such as European Directive 67/548EEC stipulates that prior to the use of plant protection products, a risk assessment has to be performed. Due to the large number of chemicals used worldwide, it is practically impossible to implement a full quantitative risk assessment for all chemicals. Therefore, chemical ranking systems can be used as an initial screening of pesticides with the view to identifying those requiring further analysis. Nineteen commonly used pesticide ranking tools were evaluated according to: scale of intervention; environmental compartments; human and ecosystem effects; and stage of development and functionality. The tools were assigned a numerical score (maximum of 15) based on their fulfilment of these criteria. This facilitated identification of tools that could be used for a given set of input data and user requirements. The results showed great diversity with total scores between 6 and 13.5. The advantages and disadvantages of each individual model are described. Care needs to be taken in selecting a pesticide ranking tool that fulfills the required criteria in terms of scale, effects, functionality, and environmental compartments analyzed.     

Association of permanent arterial hypertension with the renin-angiotensin and kinin-bradykinin system genes in children

A new approach, based on scoring, was proposed for a more objective estimation of the contribution of several genetic variants to a multifactorial disease. Two variants of the approach—genotype scoring and unfavorable factor scoring—were tested by analyzing the polymorphisms of REN (I9–83G>A), AGT (M235T), ACE (I/D), AGTR1 (1166A>C), ATGR2 (3123C>A), and BKR2 (?58T>C and I/D) in children with arterial hypertension (AH). Each of the two variants reported that the genes of the renin-angiotensin and kinin-brady-kinin systems play an important role in permanent AH in girls.     

A nonracial craniofacial perspective on human variation: A(ustralia) to Z(uni)

Dental and craniofacial measurements were collected for 57 samples from Asia, the Pacific, the aboriginal western hemisphere, and Europe. The craniofacial dimensions include many that are not obviously under the control of specific selective forces. Similar configurations for these in different samples should yield indications of recency of common ancestry according to the logic expressed by Darwin and evident in the relationships indicated by nuclear DNA comparisons. Dental dimensions, however, vary according to the length of time that different intensities in selective forces have been in operation. The craniofacial measurements were transformed into C scores and used to generate Euclidean distance dendrograms. When all the material was used to generate a single dendrogram, the European and Amerindian samples sorted into two regionally identifiable clusters, and the Asian and Pacific material sorted into the three clusters identified in separate previous studies: a Mainland Asian cluster, a Jomon-Pacific cluster and an Australo-Melanesian cluster. Since these clusters are based on variation in traits that are basically nonadaptive in nature, no hierarchical ranking is possible. The clusters simply reflect degree of relationship. This technique holds forth the promise of producing a nonracial assessment of the relationships of all the peoples of the world, past and present.     

Grades of undernutrition and socioeconomic status influence cognitive development in school children of Kolkata

Cognitive development of children is influenced by different environmental factors like nutritional and socio‐economic status. The objectives of the present study were to determine the influence of grades of undernutrition and socio‐economic status (SES) on the cognitive development of school children of Kolkata. Five hundred sixty six (566) school children having 5–12 years of age were selected from different schools of Kolkata. The cognitive development was measured by the scores of Raven's colored progressive matrices (RCPM). The chronic and acute nutritional statuses were measured from height‐for‐age (HAZ) and weight‐for‐age (WAZ) Z scores respectively with reference to the values of WHO. SES was determined by updated Kuppuswamy scale. The prevalences of undernutrition in the observed children were 57.95% (according to HAZ) and 52.82% (according to WAZ). The age dependent growth curve of RCPM scores of the observed children remains in between the 10th and 25th centile of British children. The children belonging to superior and intellectual deficit IQ classes were 21.55 and 36.40%, respectively of the total subjects. Most of the subjects belong to lower middle (39.93%) and upper middle (36.40%) class of SES. RCPM scores of school children were gradually decreased with the grades of undernutrition and SES. RCPM scores were significantly correlated with HAZ, WAZ, SES, age, and sex (P < 0.001) and strongly associated with HAZ, SES, age, and sex (P < 0.001, P < 0.05). Present study indicates that cognitive development of school children of Kolkata is influenced by the grade of undernutrition and SES. Am J Phys Anthropol 156:274–285, 2015. © 2014 Wiley Periodicals, Inc.     

Estimating Propensity Scores and Causal Survival Functions Using Prevalent Survival Data

Summary This article develops semiparametric approaches for estimation of propensity scores and causal survival functions from prevalent survival data. The analytical problem arises when the prevalent sampling is adopted for collecting failure times and, as a result, the covariates are incompletely observed due to their association with failure time. The proposed procedure for estimating propensity scores shares interesting features similar to the likelihood formulation in case‐control study, but in our case it requires additional consideration in the intercept term. The result shows that the corrected propensity scores in logistic regression setting can be obtained through standard estimation procedure with specific adjustments on the intercept term. For causal estimation, two different types of missing sources are encountered in our model: one can be explained by potential outcome framework; the other is caused by the prevalent sampling scheme. Statistical analysis without adjusting bias from both sources of missingness will lead to biased results in causal inference. The proposed methods were partly motivated by and applied to the Surveillance, Epidemiology, and End Results (SEER)‐Medicare linked data for women diagnosed with breast cancer.     

Anthocyanins of the Sterculiaceae: Flavonoid scores and Hennigian phylogenetic systematics

Published results of the distribution of anthocyanins in the Sterculiaceae have been re-interpreted on the basis of the phylogenetic status of the compounds present. A flavonoid score system was less useful than a cladistic interpretation based on Hennigian arguments.     

The effects of pasture availability for twin- and triplet-bearing ewes in mid and late pregnancy on ewe and lamb behaviour 12 to 24 h after birth

This experiment was conducted to evaluate the effect of offering ewes two different feeding levels, during mid and late pregnancy, on ewe and lamb behaviour 12 to 24 h after birth. Romney ewes, bearing twin (n = 80) or triplet foetuses (n = 56), were allocated to a pasture sward height of 2 or 4 cm between 70 and 107 days of pregnancy. In late pregnancy (day 107 to 147), half of the ewes were reallocated the alternate sward height, which produced four treatments: 2-2, 2-4, 4-2 and 4-4. Ewes were weighed on days 65, 92, 107 and 130 of pregnancy and lamb live weights were recorded 12 to 24 h after birth. Twelve to 24 h after birth the maternal behaviour score (MBS) of the ewes were determined, whilst their lambs were tagged. After the lambs were released, the behaviour of each ewe and her lambs was observed for 5 min. Ewe treatment and litter size had no effect on ewe MBS. However, as MBS increased (ewes stayed closer to lambs during tagging), ewes bleated less in a high-pitch and were quicker to make contact with their lamb. During the observation period, ewes in the 4-4 treatment had a greater percentage of their bleats in a low pitch (P < 0.05) than ewes in the 2-2 and 4-2 treatment (61.3% v. 41.3% and 38.8% low bleats, respectively) and more lambs born to 4-4 ewes (95%) bleated than lambs born to 2-2 ewes (84%; P < 0.05). However, lambs born to ewes in the 2-2 treatment bleated earlier than lambs in all other treatments (P < 0.05). Lambs born to 4-4 ewes were less likely (P < 0.05) to move towards their dam in order to make contact than lambs born to 2-2 or 4-2 ewes (3.1% v. 16.9% and 16.7%, respectively). These findings suggest that under the conditions of the present study, ewe nutrition had little effect on maternal behaviour. However, lambs born to ewes offered 2 cm pasture sward heights during mid and/or late pregnancy (2-2, 2-4 and 4-2 treatments) displayed behaviour that demonstrated greater 'need' whereas lambs born to ewes offered 4 cm during mid and late pregnancy sought less attention from their dam.     

Comparison of microsatellites and amplified fragment length polymorphism markers for parentage analysis

This study compares the properties of dominant markers, such as amplified fragment length polymorphisms (AFLPs), with those of codominant multiallelic markers, such as microsatellites, in reconstructing parentage. These two types of markers were used to search for both parents of an individual without prior knowledge of their relationships, by calculating likelihood ratios based on genotypic data, including mistyping. Experimental data on 89 oak trees genotyped for six microsatellite markers and 159 polymorphic AFLP loci were used as a starting point for simulations and tests. Both sets of markers produced high exclusion probabilities, and among dominant markers those with dominant allele frequencies in the range 0.1-0.4 were more informative. Such codominant and dominant markers can be used to construct powerful statistical tests to decide whether a genotyped individual (or two individuals) can be considered as the true parent (or parent pair). Gene flow from outside the study stand (GFO), inferred from parentage analysis with microsatellites, overestimated the true GFO, whereas with AFLPs it was underestimated. As expected, dominant markers are less efficient than codominant markers for achieving this, but can still be used with good confidence, especially when loci are deliberately selected according to their allele frequencies.     

famoz: a software for parentage analysis using dominant,codominant and uniparentally inherited markers

famoz (an acronym for father/mother) is a software useful in reconstructing parentage for dominant, codominant and uniparentally inherited markers. It is written in C and TclTk languages and is available for Unix, Linux and Windows systems at http://www.pierroton.inra.fr/genetics/labo/Software/Famoz/index.html . Parameters and assumptions used in the calculations are few and simple. Exclusion and identity probabilities, log‐likelihoods of any genetic relationship, potential father and parent or parent pair, half‐ and full‐sibship are calculated based on real or simulated data. Error rates for genotypic mistyping can be introduced. Simulations can be done to build statistical tests for parentage assignment.     

Bone microarchitecture at muscle attachment sites: The relationship between macroscopic scores of entheses and their cortical and trabecular microstructural design

The studies of entheses in bioarchaeology attempted to reconstruct the habitual physical activities of past populations. However, the studies of microarchitecture of the underlying bone are still lacking despite well‐known potential of bone internal microarchitecture to reflect mechanical loading. It is unknown whether different morphological expressions of entheseal changes (ECs) correlate with the microstructural characteristics of the underlining bone. This study analyzed bone microstructural characteristics at the entheses. Our focus was on examining the possible successive nature of the three‐stage scale of entheseal macroscopic changes by comparing EC scores with the microarchitectural features at the attachment sites. The study was based on the hypothesis that mechanical loading influences the microarchitecture of the bone at the attachment site. The bone samples were taken from 24 adult male skeletons from medieval cemeteries in Serbia, with different macroscopic expression score of EC. We evaluated the macroscopic and microscopic appearance of four entheses of the lower limbs (origin of the soleus muscle and the insertions of the adductor magnus, gluteus maximus, and iliopsoas muscles). The specimens were scanned using microcomputed tomography (Scanco µCT 40). Our data showed a lack of consistent correlation between stages of the macroscopic scoring systems with microarchitecture at the entheses, only cortical thickness was significantly different between EC stages. Analyzing relationship between trabecular and cortical bone microstructure we found correlations between cortical and trabecular variables only in Stage C. Results of our study suggest that macroscopic EC might not represent distinct successive phases in bone adaptation to mechanical loading. Am J Phys Anthropol 157:81–93, 2015. © 2014 Wiley Periodicals, Inc.     

The relationship of live animal muscular and skeletal scores, ultrasound measurements and carcass classification scores with carcass composition and value in steers

This study examined the relationship of muscular and skeletal scores and ultrasound measurements in the live animal, and carcass conformation and fat scores with carcass composition and value using 336 steers, slaughtered at 2 years of age. Live animal scores and measurements were recorded at 8 to 12 months of age and pre-slaughter. Following slaughter, each carcass was classified for conformation and fatness and the right side dissected into meat, fat and bone. Carcass conformation scores and fat scores were both measured on a continuous 15-point scale and ranged from 2.0 to 12.0 and from 2.8 to 13.3, respectively. Pre-slaughter muscular scores showed positive correlations (P < 0.001) ranging from 0.31 to 0.86 with carcass meat proportion, proportion of high-value cuts in the carcass, conformation score and carcass value, significant negative correlations with carcass fat (r = -0.13) and bone (r = -0.81) proportions, and generally low non-significant relationships with the proportion of high-value cuts in meat and carcass fat score. Pre-slaughter ultrasound muscle depth and carcass conformation score showed similar correlations with carcass traits to those using the pre-slaughter muscular scoring procedure. Pre-slaughter ultrasound fat depth showed positive correlations (P < 0.001) with carcass fat proportion (r = 0.59) and fat score (r = 0.63), and significant negative correlations (-0.23 to -0.50) with carcass meat and bone proportions, high-value cuts in the carcass and in meat, and carcass value. Pre-slaughter skeletal scores generally showed poor correlations ranging from -0.38 to 0.52 with the various carcass traits. Corresponding correlations (-0.26 to 0.44) involving records collected at 8 to 12 months of age were lower than those using pre-slaughter records. A one-unit increase in carcass conformation score increased carcass meat proportion and value by 11.2 g/kg and 5.6 cents/kg, respectively. Corresponding values for fat score were -8.2 g/kg and -5.1 cents/kg. In conclusion, both pre-slaughter live animal scores/measurements and carcass classification scores, explained an appreciable amount of the total variation in carcass meat, fat and bone proportions and carcass value, and a moderate amount of the variation in proportion of high-value meat cuts in the carcass.     

The value of muscular and skeletal scores in the live animal and carcass classification scores as indicators of carcass composition in cattle

The objective was to determine the relationship of muscular and skeletal scores taken on the live animal and carcass conformation and fat scores with carcass composition and value. Bulls (n = 48) and heifers (n = 37) of 0.75 to 1.0 late-maturing breed genotypes slaughtered at 16 and 20 months of age, respectively, were used. At 8 months of age (weaning) and immediately pre-slaughter, visual muscular scores were recorded for each animal and additionally skeletal scores were recorded pre-slaughter. Carcass weight, kidney and channel fat weight, carcass conformation and fat scores, fat depth over the longissimus dorsi muscle at the 12th (bulls) or 10th (heifers) rib and carcass length were recorded post-slaughter. Each carcass was subsequently dissected into meat, fat and bone using a commercial dissection procedure. Muscular scores taken pre-slaughter showed positive correlations with killing-out rate (r ≈ 0.65), carcass meat proportion (r ≈ 0.60), value (r ≈ 0.55) and conformation score (r ≈ 0.70), and negative correlations with carcass bone (r ≈ -0.60) and fat (r ≈ -0.4) proportions. Corresponding correlations with muscular scores at weaning were lower. Correlations of skeletal scores taken pre-slaughter, carcass length and carcass weight with killing-out rate and the various carcass traits were mainly not significant. Carcass fat depth and kidney and channel fat weight were negatively correlated with carcass meat proportion and value, and positively correlated with fat proportion. Correlations of carcass conformation score were positive (r = 0.50 to 0.68) with killing-out rate, carcass meat proportion and carcass value and negative with bone (r ≈ -0.56) and fat (r ≈ -0.40) proportions. Corresponding correlations with carcass fat score were mainly negative except for carcass fat proportion (r ≈ 0.79). A one-unit (scale 1 to 15) increase in carcass conformation score increased carcass meat proportion by 8.9 and 8.1 g/kg, decreased fat proportion by 4.0 and 2.9 g/kg and decreased bone proportion by 4.9 and 5.2 g/kg in bulls and heifers, respectively. Corresponding values per unit increase in carcass fat score were -11.9 and -9.7 g/kg, 12.4 and 9.9 g/kg, and -0.5 and -0.2 g/kg. Carcass conformation and fat scores explained 0.70 and 0.55 of the total variation in meat yield for bulls and heifers, respectively. It is concluded that live animal muscular scores, and carcass conformation and fat scores, are useful indicators of carcass meat proportion and value.     

SHA.LIN评分和S.T.O.N.E评分对经皮肾镜取石术结石清除率的预测价值比较

目的:比较SHA.LIN评分和S.T.O.N.E评分对经皮肾镜取石术结石清除率的预测价值。方法:选择我院于2016年1月-2017年12月期间行经皮肾镜取石术患者67例为研究对象,在术前对所有患者进行SHA.LIN评分和S.T.O.N.E评分。根据手术结果将患者分为结石清除组(n=49)和结石残留组(n=18),对两组患者的一般资料、SHA.LIN评分、S.T.O.N.E评分进行统计对比。采用多因素Logistic回归分析方法分析患者术后结石残留的影响因素。采用绘制ROC曲线的方法分析SHA.LIN评分和S.T.O.N.E评分对结石清除率的预测结果的敏感性和特异性。结果:67例患者术后结石清除者49例、结石残留者18例,结石清除率为73.13%。结石残留组患者手术时间、术中失血量、住院时间、结石最大截面积、最大累积截面积、结石解剖分布肾盂的发生率、受累肾盏数均高于结石清除组,穿刺通道长度低于结石清除组(P0.05)。结石清除组患者的SHA.LIN评分、S.T.O.N.E评分均低于结石残留组(P0.05)。经多因素Logistic回归分析显示,手术时间、术中失血量、SHA.LIN评分、S.T.O.N.E评分、受累肾盏数是结石残留的影响因素(P0.05)。通过绘制ROC曲线可知,SHA.LIN评分的敏感性为91.25%、特异性为89.12%、曲线下面积(AUC)为0.912(95%CI 0.869~0.948);S.T.O.N.E评分的敏感性为78.75%、特异性为84.24%、AUC为0.782(95%CI 0.690~0.871)。结论:经皮肾镜取石术患者结石清除率与SHA.LIN评分、S.T.O.N.E评分明显相关,两种评分系统均能预测患者的结石清除率,但SHA.LIN评分的敏感性、特异性高于S.T.O.N.E评分。