Explore chapters and articles related to this topic
Statistics for Genomics
Published in Altuna Akalin, Computational Genomics with R, 2020
One final method that is also popular is called the “q-value” method and related to the method above. This procedure relies on estimating the proportion of true null hypotheses from the distribution of raw p-values and using that quantity to come up with what is called a “q-value”, which is also an FDR-adjusted P-value (Storey and Tibshirani, 2003). That can be practically defined as “the proportion of significant features that turn out to be false leads.” A q-value 0.01 would mean 1% of the tests called significant at this level will be truly null on average. Within the genomics community q-value and FDR adjusted P-value are synonymous although they can be calculated differently.
Gene Expression Profiling to Detect New Treatment Targets in Leukemia and Lymphoma: A Future Perspective
Published in Gertjan J. L. Kaspers, Bertrand Coiffier, Michael C. Heinrich, Elihu Estey, Innovative Leukemia and Lymphoma Therapy, 2019
Torsten Haferlach, Wolfgang Kern, Alexander Kohlmann
SAM is available as Microsoft Excel Add-in (9). Bioconductor is an open source and open development software project for the analysis and comprehension of genomic data. Bioconductor packages provide statistical and graphical methodologies for analyzing genomic data. LIBSVM (Version 2.6) is a software solution for SVM-based classification. The q-value software takes a list of p-values resulting from the simultaneous testing of many hypotheses and estimates their q-values (10). In addition, further third party software packages can be used for statistical analyses and data visualization.
General Radiation Biophysics and Biology
Published in George W. Casarett, Radiation Histopathology, 2019
Because of the variation of RBE with dose size and/or dose rate, nature of effect, and LET, there have been chosen for practical purposes of radiation protection, single values for effectiveness of various categories of radiation quality (LET), e.g., to facilitate equating or summation of doses from various radiations in terms of effects or risks. These selected values are called “quality factors” (QF or simply Q). For conservative purposes, it has been necessary to select Q values not greatly different from RBE values that apply to various systems at various dose and dose rate levels. Consequently, Q values for high LET radiations are likely to be higher than RBE values found in many radiobiological experiments utilizing relatively high doses and/or dose rates, because of the known increase in RBE of high LET radiations with decreasing dose size and/or dose rate. The use of the RBE unit is now limited to radiobiology.
Implementing programmatic assessment transforms supervisor attitudes: An explanatory sequential mixed methods study
Published in Medical Teacher, 2021
Janica Jamieson, Margaret Hay, Simone Gibson, Claire Palermo
Supervisors engaging with the PA for the first time were emailed the survey link before and after the placement. A reminder email was sent after one week to maximise response rates. Demographic data for participants is presented as frequency or mean ± SD (range). The statement response data were entered into SPSSTM (IBM, version 26) for analysis with non-parametric tests applied. Descriptive results for statements from the two points (before and after) are given as percent agreement, representing both ‘agreed’ and ‘strongly agreed’. Participant responses were matched and analysed using the Wilcoxon signed-rank test. The Benjamini-Hochberg procedure, which controls for the false discovery rate, was applied to raw p-values, in consideration of the multiple tests conducted with a small sample. The q-values are presented with <0.05 considered statistically significant (Benjamini and Hochberg 1995). Effect size was determined using Cohen’s r equation for non-parametric methods (r= r value of 0.1 was considered a small effect, 0.3 medium, and 0.5 large (Fritz et al. 2012). Open-text survey responses were imported into NVivoTM 12 (QSR International, version 12) and analysed using the framework developed for phase two data by one author (JJ). A second author (CP) reviewed and confirmed the analysis.
The effectiveness of walking versus exercise on pain and function in chronic low back pain: a systematic review and meta-analysis of randomized trials
Published in Disability and Rehabilitation, 2019
Carla Vanti, Simone Andreatta, Silvia Borghi, Andrew Anthony Guccione, Paolo Pillastrini, Lucia Bertozzi
All effect sizes were computed using Hedges’g statistic [38]. Pro-Meta V.2.0 software (Internovi by Scarpellini Daniele s.a.s. Cesena [FC], Italy) was used for the statistical analyses. Standardized mean differences (SMDs) with 95% confidence intervals (95% CIs) were calculated for continuous data. The SMD was used because different measures were adopted by each study to address the same clinical outcome. To interpret effect size calculated with SMD, we used the method described by Cohen [39] as a guide to identify small (0.20), medium (0.50), or large (0.80) effects. Calculation of effect size was based first on the best possible data (i.e., final means, standard deviations, and sample sizes of intervention and control groups). Selected studies for which these or other crucial data were not directly reported or obtainable by contacting authors, were not included in the review. In cases where different articles covered results from the same study population, data from only one article were pooled. The Q- and I-square statistics were used to assess heterogeneity among studies [40]. A significant Q value indicates a lack of homogeneity of findings of studies. Following the approach of Higgins and Thompson [40], heterogeneity was qualified as low (25–50%), moderate (50– 75%), or high (75%). When it was possible, the potential publication bias was assessed using the Egger t test.
13C-Urea Breath Test for the Diagnosis of H. pylori Infection in Patients after Partial Gastrectomy: A Systematic Review and Meta-Analysis
Published in Journal of Investigative Surgery, 2022
Zehua Chen, Handong Liu, Yuexin Zhang, Tao Jin, Jiankun Hu, Kun Yang
The inconsistency of sensitivity and specificity indicated respectively as 81.2% and 90% and the Q value was statistically significant (P < 0.05). We can explain the heterogeneity for clinical or methodological variations. We conducted subgroup analysis by including 13C dose, postural position, and testing time. Patients were separated into groups A (100 mg) and B (75 mg) based on the dose of 13C received. The DOR of group A was 38.78, 95% CI was 14.57–102.22, I2 was 72.20, and P value was <0.01. The DOR of group B was 23.35, 95% CI was 5.56–98.02, I2 was 66.4, and P value was 0.03. On the basis of the postural position after medication, patients were separated into groups C (left decubitus group) and D (supine position). The DOR of group C was 49.68, 95% CI was 17.35–142.25, I2 was 46.50, and P value was 0.11. The DOR of group D was 34.72, 95% CI was 8.66–139.14, I2 was 0, and P value was 0.75. Based on the detection time, the patients were divided into groups E (<30 min group) and F (≥30 min group). The DOR of group E was 35.73, 95% CI was 15.81–80.73, I2 was 34.30, and P value was 0.17. The DOR of group F was 43.92, 95% CI was 5.39–357.87, I2 was 81.80, and P value was <0.01. According to location, patients were separated into groups G (Asian) and H (non-Asian). The DOR of group G was 34.31, 95% CI was 13.41–87.79, I2 was 69.30, and P value was <0.01. The DOR of group H was 55.15, 95% CI was 4.45–683.77, I2 was 67.10, and P value was 0.04.