Category: Statistics

J Clin Oncol. 2024 Apr 3:JCO2301926. doi: 10.1200/JCO.23.01926. Online ahead of print.

ABSTRACT

PURPOSE: There exists a barrier between developing and disseminating risk prediction models in clinical settings. We hypothesize that this barrier may be lifted by demonstrating the utility of these models using incomplete data that are collected in real clinical sessions, as compared with the commonly used research cohorts that are meticulously collected.

MATERIALS AND METHODS: Genetic counselors (GCs) collect family history when patients (ie, probands) come to MD Anderson Cancer Center for risk assessment of Li-Fraumeni syndrome, a genetic disorder characterized by deleterious germline mutations in the TP53 gene. Our clinical counseling-based (CCB) cohort consists of 3,297 individuals across 124 families (522 cases of single primary cancer and 125 cases of multiple primary cancers). We applied our software suite LFSPRO to make risk predictions and assessed performance in discrimination using AUC and in calibration using observed/expected (O/E) ratio.

RESULTS: For prediction of deleterious TP53 mutations, we achieved an AUC of 0.78 (95% CI, 0.71 to 0.85) and an O/E ratio of 1.66 (95% CI, 1.53 to 1.80). Using the LFSPRO.MPC model to predict the onset of the second cancer, we obtained an AUC of 0.70 (95% CI, 0.58 to 0.82). Using the LFSPRO.CS model to predict the onset of different cancer types as the first primary, we achieved AUCs between 0.70 and 0.83 for sarcoma, breast cancer, or other cancers combined.

CONCLUSION: We describe a study that fills in the critical gap in knowledge for the utility of risk prediction models. Using a CCB cohort, our previously validated models have demonstrated good performance and outperformed the standard clinical criteria. Our study suggests that better risk counseling may be achieved by GCs using these already-developed mathematical models.

PMID:38569124 | DOI:10.1200/JCO.23.01926

J Am Chem Soc. 2024 Apr 3. doi: 10.1021/jacs.3c12721. Online ahead of print.

ABSTRACT

Comparison of bonding and electronic structural features between trivalent lanthanide (Ln) and actinide (An) complexes across homologous series’ of molecules can provide insights into subtle and overt periodic trends. Of keen interest and debate is the extent to which the valence f- and d-orbitals of trivalent Ln/An ions engage in covalent interactions with different ligand donor functionalities and, crucially, how bonding differences change as both the Ln and An series are traversed. Synthesis and characterization (SC-XRD, NMR, UV-vis-NIR, and computational modeling) of the homologous lanthanide and actinide N-heterocyclic carbene (NHC) complexes [M(C₅Me₅)₂(X)(I^Me4)] {X = I, M = La, Ce, Pr, Nd, U, Np, Pu; X = Cl, M = Nd; X = I/Cl, M = Nd, Am; and I^Me4 = [C(NMeCMe)₂]} reveals consistently shorter An-C vs Ln-C distances that do not substantially converge upon reaching Am³⁺/Nd³⁺ comparison. Specifically, the difference of 0.064(6) Å observed in the La/U pair is comparable to the 0.062(4) Å difference observed in the Nd/Am pair. Computational analyses suggest that the cause of this unusual observation is rooted in the presence of π-bonding with the valence d-orbital manifold in actinide complexes that is not present in the lanthanide congeners. This is in contrast to other documented cases of shorter An-ligand vs Ln-ligand distances, which are often attributed to increased 5f vs 4f radial diffusivity leading to differences in 4f and 5f orbital bonding involvement. Moreover, in these traditional observations, as the 5f series is traversed, the 5f manifold contracts such that by americium structural studies often find no statistically significant Am³⁺vs Nd³⁺ metal-ligand bond length differences.

PMID:38569081 | DOI:10.1021/jacs.3c12721

J Appl Clin Med Phys. 2024 Apr 3:e14343. doi: 10.1002/acm2.14343. Online ahead of print.

ABSTRACT

PURPOSE: Single-isocenter multi-target intracranial stereotactic radiotherapy (SIMT) is an effective treatment for brain metastases with complex treatment plans and delivery optimization necessitating rigorous quality assurance. This work aims to assess five methods for quality assurance of SIMT treatment plans in terms of their suitability and sensitivity to delivery errors.

METHODS: Sun Nuclear ArcCHECK and SRS MapCHECK, GafChromic EBT Radiochromic Film, machine log files, and Varian Portal Dosimetry were all used to measure 15 variations of a single SIMT plan. Variations of the original plan were created with Python. They comprised various degrees of systematic MLC offsets per leaf up to 2 mm, random per-leaf variations with differing minimum and maximum magnitudes, simulated collimator, and dose miscalibrations (MU scaling). The erroneous plans were re-imported into Eclipse and plan-quality degradation was assessed by comparing each plan variation to the original clinical plan in terms of the percentage of clinical goals passing relative to the original plan. Each erroneous plan could be then ranked by the plan-quality degradation percentage following recalculation in the TPS so that the effects of each variation could be correlated with γ pass rates and detector suitability.

RESULTS & CONCLUSIONS: It was found that 2%/1 mm is a good starting point for the ArcCHECK, Portal Dosimetry, and the SRS MapCHECK methods, respectively, and provides clinically relevant error detection sensitivity. Looser dose criteria of 5%/1 mm or 5%/1.5 mm are suitable for film dosimetry and log-file-based methods. The statistical methods explored can be expanded to other areas of patient-specific QA and detector assessment.

PMID:38569013 | DOI:10.1002/acm2.14343

PLoS One. 2024 Apr 3;19(4):e0301042. doi: 10.1371/journal.pone.0301042. eCollection 2024.

ABSTRACT

BACKGROUND: Despite the well-established benefits of early initiation of breastfeeding and exclusive breastfeeding for the first six months to promote optimal neonatal and child health, evidence indicates that in Ethiopia, a significant number of newborns initiate breastfeeding late, do not adhere to exclusive breastfeeding (EBF) for the recommended duration, and instead are fed with bottles.

OBJECTIVE: To determine the proportion of delayed initiation of breast milk, exclusive breastfeeding, and its individual and community-level predictors among mothers in Ethiopia.

METHODS: A secondary data analysis was done using the 2019 Ethiopian Mini Demographic Health Survey data. We examined a weighted sample of 2,012 children born within the past 24 months and 623 children aged 0-5 months at the time of the survey. The data analysis was done using STATA version 15. To understand the variation in delayed initiation and exclusive breastfeeding, statistical measures such as the Intraclass correlation coefficient, median odds ratio, and proportional change in variance were calculated. We employed a multilevel mixed-effects logistic regression model to identify predictors for each outcome variable. Statistical significance was determined with a p-value < 0.05.

RESULTS: The proportion of delayed initiation of breast milk and exclusive breastfeeding were 24.56 and 84.5%, respectively. Women aged 34-49 years old (AOR = 0.33: 95% CI; 0.15-0.72), having a television in the house (AOR = 0.74: 95%CI; 0.33-0.97), delivered by cesarean section (AOR = 3.83: 95% CI; 1.57-9.32), and resided in the Afar regional state (AOR = 1.43: 95%CI; 1.03-12.7) were significantly associated with delayed initiation of breast milk. On the other hand, attended primary education (AOR = 0.67: 95%CI; 0.35-0.99), secondary education (AOR = 0.34: 95%CI; 0.19-0.53), women whose household headed by male (AOR = 0.68; 95% CI; 0.34-0.97), and rural residents (AOR = 1.98: 95%CI; 1.09-3.43) were significantly associated with exclusive breastfeeding practice.

CONCLUSION: Health promotion efforts that encourage timely initation of breast milk and promote EBF, focused on young mothers, those who gave birth through cesarean section, and those residing in urban and the Afar regional state. Furthermore, government health policymakers and relevant stakeholders should consider these identified predictors when revising existing strategies or formulating new policies.

PMID:38568996 | DOI:10.1371/journal.pone.0301042

Nutr Rev. 2024 Apr 3:nuae033. doi: 10.1093/nutrit/nuae033. Online ahead of print.

ABSTRACT

CONTEXT: Gestational diabetes mellitus (GDM) and preeclampsia (PE) are commonly observed medical complications in pregnancy. Dietary total fat and fatty acids associated with GDM and PE risk have been examined in several epidemiological studies. In some instances, systematic reviews and meta-analyses might provide more accurate dietary recommendations.

OBJECTIVES: This systematic review and dose-response meta-analysis was conducted to investigate the association between dietary total fat and fatty acids and the risk of GDM and PE.

DATA SOURCES: Research on dietary fat intake and the risk of GDM and PE was conducted through systematic searches of the PubMed, Scopus, and Web of Science databases for articles published up to August 19, 2023. An investigation of associations between dietary intake of total fat and fatty acids and the risk of GDM and PE was performed using prospective cohort study designs.

RESULTS: Twenty-one prospective cohort studies were considered eligible. Findings indicated that higher intakes of total fat (relative risk [RR], 1.08; 95% confidence interval [CI], 1.02-1.14), animal fat (RR, 1.56; 95%CI, 1.34-1.89), vegetable fat (RR, 1.23; 95%CI, 1.05-1.45), dietary cholesterol (RR, 1.48; 95%CI, 1.10-2.00), and omega-3 fatty acid (RR, 1.11; 95%CI, 1.02-1.20) are associated with a greater risk of GDM. However, no significant association was found between dietary total fat and fatty acids and the risk of PE. Dose-response meta-analyses suggested every 10% increment in total energy intake from total fat, 5% from animal fat, 5% from vegetable fat, and 100 mg from cholesterol was related to 15%, 12%, 7%, 14%, and 20% higher GDM risk, respectively.

CONCLUSIONS: Overall, total fat, animal fat, vegetable fat, dietary cholesterol, and omega-3 fatty acid consumption are associated with a small but statistically significant increase in GDM risk.

PROTOCOL REGISTRATION: PROSPERO (CRD42023466844).

PMID:38568994 | DOI:10.1093/nutrit/nuae033

PLoS One. 2024 Apr 3;19(4):e0297362. doi: 10.1371/journal.pone.0297362. eCollection 2024.

ABSTRACT

BACKGROUND: Toxoplasma gondii (T. gondii) is a worldwide distributed protozoan parasite which has infected a wide range of warm-blooded animals and humans. The most common form of T. gondii infection is asymptomatic (latent); nevertheless, latent toxoplasmosis can induce various alterations of sex hormones, especially testosterone, in infected humans and animals. On the other hand, testosterone is involved in behavioral traits and reproductive functions in both sexes. Hence, the purpose of this systematic review is to summarize the available evidence regarding the association between T. gondii infection and testosterone alteration.

METHODS: In the setting of a systematic review, an electronic search (any date to 10 January 2023) without language restrictions was performed using Science Direct, Web of Science, PubMed, Scopus, and Google Scholar. The PRISMA guidelines were followed. Following the initial search, a total of 12,306 titles and abstracts were screened initially; 12,281 were excluded due to the lack of eligibility criteria or duplication. Finally, 24 articles met the included criteria. A mean±standard deviation (SD) was calculated to assess the difference of testosterone between T. gondii positive and T. gondii negative humans. The possibility of publication bias was assessed using Egger’s regression. P-value < 0.05 was considered statistically significant.

RESULTS: This systematic review identified 24 articles (18 studies in humans and six studies in animals). Most human studies (13 out of 19) reported an increased level of testosterone following latent toxoplasmosis in males, while three studies reported decreased levels and two studies reported an insignificant change. Eleven articles (seven datasets in males and seven datasets in females) were eligible to be included in the data synthesis. Based on the random-effects model, the pooled mean± SD of testosterone in T. gondii positive than T. gondii negative was increased by 0.73 and 0.55 units in males and females, respectively. The Egger’s regression did not detect a statistically significant publication bias in males and females (p = value = 0.95 and 0.71), respectively. Three studies in male animals (rats, mice, and spotted hyenas) and two studies in female animals (mice and spotted hyenas) reported a decline in testosterone in infected compared with non-infected animals. While, one study in female rats reported no significant changes of testosterone in infected than non-infected animals. Moreover, two studies in male rats reported an increased level of testosterone in infected than non-infected animals.

CONCLUSIONS: This study provides new insights about the association between T. gondii infection and testosterone alteration and identifies relevant data gaps that can inform and encourage further studies. The consequence of increased testosterone levels following T. gondii infection could partly be associated with increased sexual behavior and sexual transmission of the parasite. On the other hand, declining testosterone levels following T. gondii infection may be associated with male reproductive impairments, which were observed in T. gondii-infected humans and animals. Furthermore, these findings suggest the great need for more epidemiological and experimental investigations in depth to understand the relationship between T. gondii infection and testosterone alteration alongside with future consequences of testosterone alteration.

PMID:38568993 | DOI:10.1371/journal.pone.0297362

J Agric Food Chem. 2024 Apr 3. doi: 10.1021/acs.jafc.3c08140. Online ahead of print.

ABSTRACT

A global demand for tea tree oil (TTO) has resulted in increased adulteration in commercial products. In this study, we use a novel enantiomeric gas chromatography mass spectrometry method for chiral analysis of key terpenes ((±)-terpinen-4-ol, (±)-α-terpineol, and (±)-limonene) and quantification of components present at >0.01% to test different methods of identifying adulterated TTO. Data from authentic Australian (n = 88) and oxidized (n = 12) TTO samples of known provenance were consistent with recommended ranges in ISO 4730:2017 and previously published enantiomeric ratios, with p-cymene identified as the major marker of TTO oxidation. The 15 ISO 4730:2017 constituents comprised between 84.5 and 89.8% of the total ion chromatogram (TIC) peak area. An additional 53 peaks were detected in all samples (7.3-11.0% of TIC peak area), while an additional 43 peaks were detected in between 0 and 99% (0.15-2.0% of the TIC peak area). Analysis of nine commercial samples demonstrated that comparison to the ISO 4730:2017 standard does not always identify adulterated TTO samples. While statistical analysis of minor components in TTO did identify two commercial samples that differed from authentic TTO, the (+)-enantiomer percentages for limonene, terpinen-4-ol, and α-terpineol provided clearer evidence that these samples were adulterated. Thus, straightforward identification of unadulterated and unoxidized TTO could be based on analysis of appropriate enantiomeric ratios and quantitation of the p-cymene percentage.

PMID:38568986 | DOI:10.1021/acs.jafc.3c08140

PLoS One. 2024 Apr 3;19(4):e0301103. doi: 10.1371/journal.pone.0301103. eCollection 2024.

ABSTRACT

Birch bark tar is the most widely documented adhesive in prehistoric Europe. More recent periods attest to a diversification in terms of the materials used as adhesives and their application. Some studies have shown that conifer resins and beeswax were added to produce compound adhesives. For the Iron Age, no comparative large-scale studies have been conducted to provide a wider perspective on adhesive technologies. To address this issue, we identify adhesive substances from the Iron Age in north-eastern France. We applied organic residue analysis to 65 samples from 16 archaeological sites. This included residues adhering to ceramics, from vessel surface coatings, repaired ceramics, vessel contents, and adhesive lumps. Our findings show that, even during the Iron Age in north-eastern France, birch bark tar is one of the best-preserved adhesive substances, used for at least 400 years. To a lesser extent, Pinaceae resin and beeswax were also identified. Through statistical analyses, we show that molecular composition differs in samples, correlating with adhesive function. This has implications for our understanding of birch bark tar production, processing and mode of use during the Iron Age in France and beyond.

PMID:38568980 | DOI:10.1371/journal.pone.0301103

Proc Natl Acad Sci U S A. 2024 Apr 9;121(15):e2322083121. doi: 10.1073/pnas.2322083121. Epub 2024 Apr 3.

ABSTRACT

While reliable data-driven decision-making hinges on high-quality labeled data, the acquisition of quality labels often involves laborious human annotations or slow and expensive scientific measurements. Machine learning is becoming an appealing alternative as sophisticated predictive techniques are being used to quickly and cheaply produce large amounts of predicted labels; e.g., predicted protein structures are used to supplement experimentally derived structures, predictions of socioeconomic indicators from satellite imagery are used to supplement accurate survey data, and so on. Since predictions are imperfect and potentially biased, this practice brings into question the validity of downstream inferences. We introduce cross-prediction: a method for valid inference powered by machine learning. With a small labeled dataset and a large unlabeled dataset, cross-prediction imputes the missing labels via machine learning and applies a form of debiasing to remedy the prediction inaccuracies. The resulting inferences achieve the desired error probability and are more powerful than those that only leverage the labeled data. Closely related is the recent proposal of prediction-powered inference [A. N. Angelopoulos, S. Bates, C. Fannjiang, M. I. Jordan, T. Zrnic, Science 382, 669-674 (2023)], which assumes that a good pretrained model is already available. We show that cross-prediction is consistently more powerful than an adaptation of prediction-powered inference in which a fraction of the labeled data is split off and used to train the model. Finally, we observe that cross-prediction gives more stable conclusions than its competitors; its CIs typically have significantly lower variability.

PMID:38568975 | DOI:10.1073/pnas.2322083121

PLoS One. 2024 Apr 3;19(4):e0299993. doi: 10.1371/journal.pone.0299993. eCollection 2024.

ABSTRACT

The selection of proper reference genes is critical for accurate gene expression analysis in all fields of biological and medical research, mainly because there are many distinctions between different tissues and specimens. Given this variability, even in known classic reference genes, demands of a comprehensive analysis platform is needed to identify the most suitable genes for each study. For this purpose, we present an analysis tool for assisting in decision-making in the analysis of reverse transcription-quantitative polymerase chain reaction (RT-qPCR) data. EndoGeneAnalyzer, an open-source web tool for reference gene analysis in RT-qPCR studies, was used to compare the groups/conditions under investigation. This interactive application offers an easy-to-use interface that allows efficient exploration of datasets. Through statistical and stability analyses, EndoGeneAnalyzer assists in the select of the most appropriate reference gene or set of genes for each condition. It also allows researchers to identify and remove unwanted outliers. Moreover, EndoGeneAnalyzer provides a graphical interface to compare the evaluated groups, providing a visually informative differential analysis.

PMID:38568963 | DOI:10.1371/journal.pone.0299993