Statistical evaluation of differential expression on cDNA nylon arrays with replicated experiments

In this paper we focus on the detection of differentially expressed genes according to changes in hybridization signals using statistical tests. These tests were applied to 14 208 zebrafish cDNA clones that were immobilized on a nylon support and hybridized with radioactively labeled target mRNA from wild-type and lithium-treated zebrafish embryos. The methods were evaluated with respect to 16 control clones that correspond to eight different genes which are known to be involved in dorso-ventral axis specification. Moreover, 4608 Arabidopsis thaliana clones on the same array were used to judge statistical significance of expression changes and to control the false positive rates of the test decisions. Utilizing this special array design we show that differential expression of a high proportion of cDNA clones (15/16) and the respective genes (7/8) were identified, with a false positive error of <5% using the constant control data. Furthermore, we investigated the influence of the number of repetitions of experiments on the accuracy of the procedures with experimental and simulated data. Our results suggest that the detection of differential expression with repeated hybridization experiments is an accurate and sensitive way of identifying even small expression changes (1:1.5) of a large number of genes in parallel.     

HOW TO ESTIMATE AND USE THE VARIANCE OF d' FROM DIFFERENCE TESTS

d' is an estimate of δ, a measure of the degree of sensory difference between two products, that can be obtained easily using tables, from the proportion of difference tests performed correctly. Tables of δ are available for the 2-AFC, 3-AFC, triangular and duo-trio tests. Tables for calculating the variance of d' for these tests are provided in this paper. They can be used for comparison of d's, especially for those obtained from different difference tests. A simple procedure is described here for computing values for the variance of d'. Having obtained the variance, confidence intervals for d' can be obtained, tests of significance for d' can be made as well as tests of whether two or more d's are significantly different. The formula and tables for the number of judgments required for the estimation of δ are given also in this paper.     

EVALUATION OF OLFACTORY INTENSITY: COMPARATIVE STUDY OF TWO METHODS

Two experimental procedures recommended for the evaluation of the psychophysical characteristics of odorous compounds, olfactory matching with the 1-butanol scale and cross-modality matching with the finger span are compared. The intensity of ethyl butyrate and guaiacol solutions presented at four different concentration levels was evaluated by a panel of sixteen subjects over five repetitions using the two methods. Each stimulus was delivered to the subject from a Teflon bag through a nose-shaped glass sniffing port.
The discrimination ability, repeatability, panel homogeneity and within-subject variability of the methods were assessed. Results indicate that with both methods, subjects were able to highly discriminate the presented concentration levels of the odorants. There were no great difference in repeatability and the same individual variability was observed between both methods. However, the smaller within-subject variability highlighted for the 1-butanol scale method suggested that this method is potentially more powerful than finger-span method.     

Linkage analysis in the presence of errors IV: joint pseudomarker analysis of linkage and/or linkage disequilibrium on a mixture of pedigrees and singletons when the mode of inheritance cannot be accurately specified

There is a lot of confusion in the literature about the "differences" between "model-based" and "model-free" methods and about which approach is better suited for detection of the genes predisposing to complex multifactorial phenotypes. By starting from first principles, we demonstrate that the differences between the two approaches have more to do with study design than statistical analysis. When simple data structures are repeatedly ascertained, no assumptions about the genotype-phenotype relationship need to be made for the analysis to be powerful, since simple data structures admit only a small number of df. When more complicated and/or heterogeneous data structures are ascertained, however, the number of df in the underlying probability model is too large to have a powerful, truly "model-free" test. So-called "model-free" methods typically simplify the underlying probability model by implicitly assuming that, in some sense, all meioses connecting two affected individuals are informative for linkage with identical probability and that the affected individuals in a pedigree share as many disease-predisposing alleles as possible. By contrast, "model-based" methods add structure to the underlying parameter space by making assumptions about the genotype-phenotype relationship, making it possible to probabilistically assign disease-locus genotypes to all individuals in the data set on the basis of the observed phenotypes. In this study, we demonstrate the equivalence of these two approaches in a variety of situations and exploit this equivalence to develop more powerful and efficient likelihood-based analogues of "model-free" tests of linkage and/or linkage disequilibrium. Through the use of a "pseudomarker" locus to structure the space of observations, sib-pairs, triads, and singletons can be analyzed jointly, which will lead to tests that are more well-behaved, efficient, and powerful than traditional "model-free" tests such as the affected sib-pair, transmission/disequilibrium, haplotype relative risk, and case-control tests. Also described is an extension of this approach to large pedigrees, which, in practice, is equivalent to affected relative-pair analysis. The proposed methods are equally applicable to two-point and multipoint analysis (using complex-valued recombination fractions).     

Optimal replication and the importance of experimental design for gel-based quantitative proteomics

Quantitative proteomic studies, based on two-dimensional gel electrophoresis, are commonly used to find proteins that are differentially expressed between samples or groups of samples. These proteins are of interest as potential diagnostic or prognostic biomarkers, or as proteins associated with a trait. The complexity of proteomic data poses many challenges, so while experiments may reveal proteins that are differentially expressed, these are often not significant when subjected to rigorous statistical analysis. However, this can be addressed through appropriate experimental design. A good experimental design considers the impact of different sources of variation, both analytical and biological, on the statistical importance of the results. The design should address the number of samples that must be analyzed and the number of replicate gels per sample, in the context of a particular minimum difference that one is seeking to achieve. In this study, we explore the ways to improve the quality of protein expression data from 2-DE gels, and describe an approach for defining the number of samples required and the number of gels per sample. It has been developed for the simplest of situations, two groups of samples with variation at two levels: between samples and between gels. This approach will also be useful as a guide for more complex designs involving more than two groups of samples. We describe some Internet-accessible tools that can assist in the design of proteomic studies.     

Calibration and evaluation of field cage for oviposition study with Aedes (Stegomyia) aegypti female (L.) (Diptera: Culicidae)

Differences among results gathered from insect behavior studies conducted in laboratory and field situations are due to ambient variables that differ greatly between both environments. In laboratory studies the environmental conditions can be controlled whereas in field temperature, humidity and air velocity vary uncontrollably. The objective of this study was to calibrate and evaluate an experimental area (field cage) (14 x 7 x 3.5 m) subdivided into eight test cages (2.5 x 2.5 x 2 m) for use in behavioral oviposition tests of Aedes aegypti (L.) mosquitoes for developing a new methodology to assess attractants and oviposition traps. Test cage calibration involved: (1) minimal experiment duration tests; (2) optimal female release number per traps test and (3) trap placement tests. All tests used gravid A. aegypti females; 3-4 days post blood meal and the sticky trap MosquiTRAP to catch adults. Ninety percent of the females released were recaptured 2h after the beginning of the experiment, and this allowed up to 32 test repetitions/day to be conducted in the field cage. The minimum number of females necessary to conduct statistical analyses was 20 females/trap/test per cage. No significant difference was found in the behavioral response of gravid females to four different trap positions within test cages. Field trapping results with attractant were similar to those in the field cage. Therefore, the field cage could replace field trapping for evaluating at least mosquito traps and oviposition attractants for A. aegypti.     

The effects of short-term unilateral and bilateral lower-body resistance training on measures of strength and power

The purpose of this study was to compare the effects of short-term unilateral resistance training (UL) and bilateral resistance training (BL) with free weights on several tests of unilateral and bilateral lower-body strength and power in men and women. Thirty-eight untrained men and women (mean body mass 78.3 +/- 21.47 kg; age 20.74 +/- 2.6 years) completed the study. The groups trained 2 days per week for 8 weeks with free weights and 2 days per week for 5 of the 8 weeks with plyometric drills. The resistance-training program consisted of a progression from 3 sets of 15 repetitions at 50% of the subject's predicted 1 repetition maximum (1RM) to 6 sets of 5 repetitions at 87% 1RM. Training volume and intensity were equal for each group. The free-weight squat was used to measure unilateral and bilateral strength. Power was measured by the Magaria-Kalamen stair-climb test and the unilateral and bilateral vertical jump test. Analysis of covariance was used to analyze differences between men and women and the interaction of group and gender. Pretest scores were used as the covariate. The UL group improved more than the BL group on the unilateral vertical jump height (p = 0.001) and relative power (p = 0.013). After adjusting for pretest differences, the improved scores on all tests, except for the unilateral squat, were similar between the men and the women. No significant interactions on all tests were found for the men or women comparison between training groups. These results indicate that UL and BL are equally effective for early phase improvement of unilateral and bilateral leg strength and power in untrained men and women.     

Prediction of muscle fiber composition using multiple repetition testing

Direct determination of muscle fiber composition is invasive and expensive, with indirect methods also requiring specialist resources and expertise. Performing resistance exercises at 80% 1RM is suggested as a means of indirectly estimating muscle fiber composition, though this hypothesis has never been validated against a direct method. The aim of the study was to investigate the relationship between the number of completed repetitions at 80% 1RM of back squat exercise and muscle fiber composition. Thirty recreationally active participants’ (10 females, 20 males) 1RM back squat load was determined, before the number of consecutive repetitions at 80% 1RM was recorded. The relationship between the number of repetitions and the percentage of fast-twitch fibers from vastus lateralis was investigated. The number of completed repetitions ranged from 5 to 15 and was independent of sex, age, 1RM, training frequency, training type, training experience, BMI or muscle fiber cross-sectional area. The percentage of fast-twitch muscle fibers was inversely correlated with the number of repetitions completed (r = –0.38, P = 0.039). Participants achieving 5 to 8 repetitions (n = 10) had significantly more fast-twitch muscle fibers (57.5 ± 9.5 vs 44.4 ± 11.9%, P = 0.013) than those achieving 11–15 repetitions (n = 11). The remaining participants achieved 9 or 10 repetitions (n = 9) and on average had equal proportion of fast- and slow-twitch muscle fibers. In conclusion, the number of completed repetitions at 80% of 1RM is moderately correlated with muscle fiber composition.     

Moderate resistance training volume produces more favorable strength gains than high or low volumes during a short-term training cycle

The purpose of this study was to examine the effects of 3 resistance training volumes on maximal strength in the snatch (Sn), clean & jerk (C&J), and squat (Sq) exercises during a 10-week training period. Fifty-one experienced (>3 years), trained junior lifters were randomly assigned to one of 3 groups: a low-volume group (LVG, n = 16), a moderate-volume group (MVG, n = 17), and a high-volume group (HVG, n = 18). All subjects trained 4-5 days a week with a periodized routine using the same exercises and relative intensities but a different total number of sets and repetitions at each relative load: LVG (1,923 repetitions), MVG (2,481 repetitions), and HVG (3,030 repetitions). The training was periodized from moderate intensity (60- 80% of 1 repetition maximum [1RM]) and high number of repetitions per set (2-6) to high intensity (90-100% of 1RM) and low number of repetitions per set (1-3). During the training period, the MVG showed a significant increase for the Sn, C&J, and Sq exercises (6.1, 3.7, and 4.2%, respectively, p < 0.01), whereas in the LVG and HVG, the increase took place only with the C&J exercise (3.7 and 3%, respectively, p < 0.05) and the Sq exercise (4.6%, p < 0.05, and 4.8%, p < 0.01, respectively). The increase in the Sn exercise for the MVG was significantly higher than in the LVG (p = 0.015). Calculation of effect sizes showed higher strength gains in the MVG than in the HVG or LVG. There were no significant differences between the LVG and HVG training volume-induced strength gains. The present results indicate that junior experienced lifters can optimize performance by exercising with only 85% or less of the maximal volume that they can tolerate. These observations may have important practical relevance for the optimal design of strength training programs for resistance-trained athletes, since we have shown that performing at a moderate volume is more effective and efficient than performing at a higher volume.     

Simple test statistics for major gene detection: a numerical comparison

Summary We compare 22 simple tests for the detection of major gene segregation in livestock populations. These tests belong to two groups: methods based on the comparison of within-family distribution and methods based on the comparison of parents' and offspring performances. The power of the 22 tests and the robustness of the two more powerful of these 22 are evaluated by simulation. Thirteen types of major loci, differing in the within-genotype means, variances or alleles frequencies, are studied. Thirty hierarchically balanced populations defined by the number of sire families (5–20), dams per sire (1–20) and progenies per dam (1–20) are simulated. The quantiles are estimated from 2000 samples, the power from 1000 samples and the robustness from 100 samples. The more powerful tests are the within family-variance heterogenity test (Bartlett test) and the within-family mean-variance regression (Fain 1978). Their robustness may be very low, in particular when the trait distribution is skewed.     

A reappraisal of Bateman's classic study of intrasexual selection

Bateman's (1948) study showing greater variances in number of mates and reproductive success in male than female Drosophila melanogaster is a foundational paper in sexual selection. Here we show for the first time that his methods had flaws, including the elimination of genetic variance, sampling biases, miscalculations of fitness variances, statistical pseudo-replication, and selective presentation of data. We conclude that Bateman's results are unreliable, his conclusions are questionable, and his observed variances are similar to those expected under random mating. Despite our analysis, we do not intend this article as a criticism of Bateman; he accomplished his work without modern computational tools, and his approach was groundbreaking emphasizing the significance of fitness variance for sexual selection. However, this reanalysis has implications for what counts as evidence for sexual selection and we believe that our concerns should be of interest to contemporary students of sexual selection. We call for repetitions of Bateman's study using modern statistical and molecular methods.     

The effect of different rest intervals between sets on volume components and strength gains

The purpose of this study was to compare squat strength gains and volume components when resting 2 minutes vs. 4 minutes between sets over multiple mesocycles. After the first squat 1 repetition maximum, 15 trained men were matched and randomly assigned to either a 2-minute (n = 7) or a 4-minute (n = 8) rest interval group. Each group performed the same training program, with the only difference being the length of the rest interval between sets. Subjects performed two squat workouts per week; one was labeled as Heavy and the other was labeled as Light. The squat workouts varied in the intensity, number of sets, and repetitions performed per set in a nonlinear periodized manner throughout each mesocycle. Differences in strength gains and volume components (the load utilized per set, the repetitions performed per set, the intensity per set, and the volume performed per workout) were compared between groups. Both groups demonstrated large strength gains; however, these differences were not significant between groups (P = 0.47). During all mesocycles, the 4-minute group demonstrated significantly higher total volumes for the Heavy workouts (P < 0.05). The findings of the present study indicate that large squat strength gains can be achieved with a minimum of 2 minutes' rest between sets, and little additional gains are derived from resting 4 minutes between sets. Therefore, athletes attempting to achieve specific volume goals may need longer rest intervals initially but may later adapt so that shorter rest intervals can be utilized without excessive fatigue, leaving additional time to focus on other conditioning priorities.     

用概率单位-平行线法和常规法统计分析几种疫苗效力检测结果的比较

本文采用概率转换-平行线法并应用微机程序对乙肝疫苗30批、狂犬疫苗34批双百日咳菌苗30批的效力试验结果进行了统计分析,并与现行常现统计法即Reed-Muench和Wilsonworcester法统计结果作了比较。 1.采用P-P法统计分析乙肝疫苗、狂犬病疫苗及百日咳菌苗效力试验结果,经对回归方程的线性及平行性检验成立的批数分别占83.3％、82.3％和83.3％,结果表明该法可以检查出由于动物差异及实验误差导致结果不成立的批数,应予复试。 2.P-P法与常规法(R-M法及W-W法)统计分析给果对三种制品评价一致的符合率分别为80％、79.4％和76.6％,两会判定结果一致,但前者更有科学性。 3.由P-P法统计分析结果,三种制品都有部分批的95％可信区间上下限值差距较大,该差距来源于动物差异及实验误差,说明在检测试验中,必需按规程要求严格筛选适宜动物和严格试验操作,以减少实验误差。对效力值可信区间应控制何种适宜范围,尚需作进一步探讨。     

Beyond Bonferroni: less conservative analyses for conservation genetics

Studies in conservation genetics often attempt to determine genetic differentiation between two or more temporally or geographically distinct sample collections. Pairwise p-values from Fisher’s exact tests or contingency Chi-square tests are commonly reported with a Bonferroni correction for multiple tests. While the Bonferroni correction controls the experiment-wise α, this correction is very conservative and results in greatly diminished power to detect differentiation among pairs of sample collections. An alternative is to control the false discovery rate (FDR) that provides increased power, but this method only maintains experiment-wise α when none of the pairwise comparisons are significant. Recent modifications to the FDR method provide a moderate approach to determining significance level. Simulations reveal that critical values of multiple comparison tests with both the Bonferroni method and a modified FDR method approach a minimum asymptote very near zero as the number of tests gets large, but the Bonferroni method approaches zero much more rapidly than the modified FDR method. I compared pairwise significance from three published studies using three critical values corresponding to Bonferroni, FDR, and modified FDR methods. Results suggest that the modified FDR method may provide the most biologically important critical value for evaluating significance of population differentiation in conservation genetics.␣Ultimately, more thorough reporting of statistical significance is needed to allow interpretation of biological significance of genetic differentiation among populations.An erratum to this article can be found at     

A fast, stochastic threading algorithm for proteins

MOTIVATION: Sequences for new proteins are being determined at a rapid rate, as a result of the Human Genome Project, and related genome research. The ability to predict the three-dimensional structure of proteins from sequence alone would be useful in discovering and understanding their function. Threading, or fold recognition, aims to predict the tertiary structure of a protein by aligning its amino acid sequence with a large number of structures, and finding the best fit. This approach depends on obtaining good performance from both the scoring function, which simulates the free energy for given trial alignments, and the threading algorithm, which searches for the lowest-score alignment. It appears that current scoring functions and threading algorithms need improvement. RESULTS: This paper presents a new threading algorithm. Numerical tests demonstrate that it is more powerful than two popular approximate algorithms, and much faster than exact methods.     

Increased number of forced repetitions does not enhance strength development with resistance training

Some research suggests that strength improvements are greater when resistance training continues to the point at which the individual cannot perform additional repetitions (i.e., repetition failure). Performing additional forced repetitions after the point of repetition failure and thus further increasing the set volume is a common resistance training practice. However, whether short-term use of this practice increases the magnitude of strength development with resistance training is unknown and was investigated here. Twelve basketball and 10 volleyball players trained 3 sessions per week for 6 weeks, completing either 4 x 6, 8 x 3, or 12 x 3 (sets x repetitions) of bench press per training session. Compared with the 8 x 3 group, the 4 x 6 protocol involved a longer work interval and the 12 x 3 protocol involved higher training volume, so each group was purposefully designed to elicit a different number of forced repetitions per training session. Subjects were tested on 3- and 6-repetition maximum (RM) bench press (81.5 +/- 9.8 and 75.9 +/- 9.0 kg, respectively, mean +/- SD), and 40-kg Smith Machine bench press throw power (589 +/- 100 W). The 4 x 6 and 12 x 3 groups had more forced repetitions per session (p < 0.01) than did the 8 x 3 group (4.1 +/- 2.6, 3.1 +/- 3.5, and 1.2 +/- 1.8 repetitions, respectively), whereas the 12 x 3 group performed approximately 40% greater work and had 30% greater concentric time. As expected, all groups improved 3RM (4.5 kg, 95% confidence limits, 3.1- 6.0), 6RM (4.7 kg, 3.1-6.3), bench press throw peak power (57 W, 22-92), and mean power (23 W, 4-42) (all p < or = 0.02). There were no significant differences in strength or power gains between groups. In conclusion, when repetition failure was reached, neither additional forced repetitions nor additional set volume further improved the magnitude of strength gains. This finding questions the efficacy of adding additional volume by use of forced repetitions in young athletes with moderate strength training experience.     

A modified YMCA bench press test as a predictor of 1 repetition maximum bench press strength

This study evaluated the influence of cadence on the Young Men's Christian Association (YMCA) bench press test for predicting 1 repetition maximum (1RM) bench press test performance. Fifty-eight medical students (37 men, 21 women) were evaluated for anthropometric variables (age, height, weight, fat-free mass, and percent fat), 1RM bench press, and 2 cadence tests of muscular endurance performed at cadences of 30 and 60 repetitions per minute (reps.min(-1)). Each test was performed on a separate day, with 5 days rest in between. There was no significant difference among the number of repetitions performed at each cadence by men, whereas women performed significantly more repetitions at the slower cadence. Repetitions at either cadence were good predictors of 1RM bench press in both genders (men: 30 reps.min(-1), r(2) = 0.757, standard error of the estimate [SEE] = 8.0 kg; 60 reps.min(-1), r(2) = 0.884, SEE = 8.2 kg; women: 30 reps.min(-1), r(2) = 0.754, SEE = 3.1 kg; 60 reps.min(-1), r(2) = 0.816, SEE = 2.7 kg). The addition of anthropometric dimensions to the regression equations did not improve predictive accuracy. Using both fast and slow cadences, the YMCA bench press test can provide a valid estimation of 1RM performance in untrained young men and women.     

A THEORETICAL NOTE ON DIFFERENCE TESTS: MODELS, PARADOXES AND COGNITIVE STRATEGIES

A simple nonmathematical introduction is given to Thurstonian modeling and its application to difference testing. The theoretical treatment explains differences in performance noted for various difference test protocols and the so-called paradox of discriminatory nondiscriminators, whereby a slight change in instructions to the judge can alter the proportion of tests performed correctly. From this, the assumptions in the use of binomial statistics for analyzing difference tests and their shortcomings, are discussed. New ideas on the generality of the paradox of discriminatory nondiscriminators are discussed, along with the effects of the cognitive strategy adopted by the judge during testing.