Empirical investigations into Kruskal-Wallis power studies utilizing Bernstein fits, simulations and medical study datasets

Clark, Jeremy S. C.; Kulig, Piotr; Podsiadło, Konrad; Rydzewska, Kamila; Arabski, Krzysztof; Białecka, Monika; Safranow, Krzysztof; Ciechanowicz, Andrzej

doi:10.1038/s41598-023-29308-2

Cited by 10 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consequently, the total number of runs amounted to 36, with a total of 108 validations from GP. The average accuracy of validations and testing accuracy from the three split ratio executions were recorded for further analysis using the Kruskal-Wallis statistical significance test [23,24]. Figure 2 depicts the Python code for implementing TPOT, a simple and efficient GP automated machine learning tool.…”

Section: Experimental Methodsmentioning

confidence: 99%

Tree-Based Pipeline Optimization Machine Learning in Classifying Whistleblowing of Academic Misconduct

Rahayu Abdul Rahman,

Suraya Masrom,

Jihadah Ahmad

et al. 2024

ARASET

View full text Add to dashboard Cite

The critical issue of academic misconduct is of utmost importance in the field of education and understanding whistleblowing behaviour can be a potential measure to effectively address this issue. This paper highlights the benefits of using the Tree-based Pipeline Optimization (TPOT) framework as a user-friendly tool for implementing machine learning techniques in studying whistleblowing behaviour among students in universities in Indonesia and Malaysia. The paper demonstrates the ease of implementing TPOT, making it accessible to inexpert computing scientists, and showcases highly promising results from the whistleblowing classification models trained with TPOT. Performance metrics such as Area Under Curve (AUC) are used to measure the reliability of the TPOT framework, with some models achieving AUC values above 90%, and the best AUC was 99% by TPOT with a Genetic Programming population size of 40. The paper’s main contribution lies in the empirical demonstration and findings that resulted in achieving the optimal outcomes from the whistleblowing case study. This paper sheds light on the potential of TPOT as an easy and rapid implementation tool for AI in the field of education, addressing the challenges of academic misconduct and showcasing promising results in the context of whistleblowing classification.

show abstract

Section: Experimental Methodsmentioning

confidence: 99%

Tree-Based Pipeline Optimization Machine Learning in Classifying Whistleblowing of Academic Misconduct

Rahayu Abdul Rahman,

Suraya Masrom,

Jihadah Ahmad

et al. 2024

ARASET

View full text Add to dashboard Cite

show abstract

“…The effect sizes for these (from Sobalska-Kwapis et al 14 ) were: rs1558902: beta = 0.349, 95% CI = 0.189:0.509, p = 1.93x10 − 5 ; rs1421085: beta = 0.345, 95% CI = 0.185:0.505, p = 2.42x10 − 5 ; and rs9939609: beta = 0.312, 95% CI = 0.152:0.472, p = 1.33x10 − 4 . From the actual SNP values for these three SNPs, three power estimations were performed using a Monte-Carlo Kruskal-Wallis power tool 22 (see Supplemental_File_S5) which gave estimated power of 85.5%, 86.6% and 82.1%, respectively.…”

Section: Methodsmentioning

confidence: 99%

ERV3-1/ZNF117: rs67047829 association with substantial protection against obesity

Clark

Podsiadło

Sobalska‐Kwapis

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

There is now substantial evidence that zinc finger proteins are implicated in adiposity. High-frequency pretermination codons (PTCs) confer near-neutral selection. Aims were to datamine for high-frequency-PTC single nucleotide polymorphisms (SNPs; n = 141; one linked with ERV3-1/ZNF117) from a database with > 550 000 variants and analyze possible association with obesity in a large Polish sample (n = 5757). Body mass index (BMI) was regressed (males/females together or separately) against genetic models. Stringent regression for rs67047829 uncovered an interaction-independent significant association between this high-frequency PTC-SNP and BMI with both sexes together: mean BMI ± standard deviation (n): GG, 25.4 ± 4.59 (3650), GA, 25.0 ± 4.28 (731); AA, 23.4 ± 3.60 (44); additive model adjusted for age and sex: p = 4.08x10− 5; beta: -0.0458, 95% confidence interval (CI): -0.0732:-0.0183; surviving Bonferroni correction; and with males: GG, 24.8 ± 4.94 (1878); GA, 24.2 ± 4.31 (386); AA, 22.4 ± 3.69 (23); p = 4.20x10− 4; beta: -0.0573, CI: -0.0947:-0.0199. For average-height males the difference between GG and AA genotypes would correspond to ~ 6 kg, suggesting considerable protection against obesity. rs67047829 is a PTC-SNP in ERV3-1 which lies upstream of, and shares an exonic region and possibly a promoter with, ZNF117, previously associated with adiposity and type 2 diabetes. As this result occurs in a near-neutral Mendelian setting, a drug target involving ERV3-1/ZNF117 potentially might provide considerable benefits with minimal side-effects. This result needs to be replicated, followed by analysis of splice-variant mRNA and protein expression.

show abstract

“…The effect sizes for these (from Sobalska-Kwapis et al 14 ) were: rs1558902: beta = 0.349, 95% CI 0.189:0.509, p = 1.93 × 10 –5 ; rs1421085: beta = 0.345, 95% CI 0.185:0.505, p = 2.42 × 10 –5 ; and rs9939609: beta = 0.312, 95% CI 0.152:0.472, p = 1.33 × 10 –4 . From the actual SNP values for these three SNPs, three power estimations were performed using our Monte-Carlo Kruskal–Wallis power tool 22 (see Supplementary File S5 ) which gave estimated power of 85.5%, 86.6% and 82.1%, respectively.…”

Section: Methodsmentioning

confidence: 99%

rs67047829 genotypes of ERV3-1/ZNF117 are associated with lower body mass index in the Polish population

Clark,

Podsiadło,

Sobalska-Kwapis

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

There is now substantial evidence that zinc-finger proteins are implicated in adiposity. Aims were to datamine for high-frequency (near-neutral selection) pretermination-codon (PTC) single-nucleotide polymorphisms (SNPs; n = 141) from a database with > 550,000 variants and analyze possible association with body mass index in a large Polish sample (n = 5757). BMI was regressed (males/females together or separately) against genetic models. Regression for rs67047829 uncovered an interaction-independent association with BMI with both sexes together: mean ± standard deviation, kg/m2: [G];[G], 25.4 ± 4.59 (n = 3650); [G](;)[A], 25.0 ± 4.28 (n = 731); [A];[A], 23.4 ± 3.60 (n = 44); additive model adjusted for age and sex: p = 4.08 × 10–5; beta: − 0.0458, 95% confidence interval (CI) − 0.0732 : − 0.0183; surviving Bonferroni correction; for males: [G];[G], 24.8 ± 4.94 (n = 1878); [G](;)[A], 24.2 ± 4.31 (n = 386); [A];[A], 22.4 ± 3.69 (n = 23); p = 4.20 × 10–4; beta: − 0.0573, CI − 0.0947 : − 0.0199. For average-height males the difference between [G];[G] and [A];[A] genotypes would correspond to ~ 6 kg, suggesting considerable protection against increased BMI. rs67047829 gives a pretermination codon in ERV3-1 which shares an exonic region and possibly promoter with ZNF117, previously associated with adiposity and type-2 diabetes. As this result occurs in a near-neutral Mendelian setting, a drug targetting ERV3-1/ZNF117 might potentially provide considerable benefits with minimal side-effects. This result needs to be replicated, followed by analyses of splice-variant mRNAs and protein expression.

show abstract

Empirical investigations into Kruskal-Wallis power studies utilizing Bernstein fits, simulations and medical study datasets

Cited by 10 publications

References 10 publications

Tree-Based Pipeline Optimization Machine Learning in Classifying Whistleblowing of Academic Misconduct

Tree-Based Pipeline Optimization Machine Learning in Classifying Whistleblowing of Academic Misconduct

ERV3-1/ZNF117: rs67047829 association with substantial protection against obesity

rs67047829 genotypes of ERV3-1/ZNF117 are associated with lower body mass index in the Polish population

Contact Info

Product

Resources

About