Tag: statistics

Commun Biol. 2022 Oct 6;5(1):1061. doi: 10.1038/s42003-022-03978-6.

ABSTRACT

The contribution of germline copy number variants (CNVs) to risk of developing cancer in individuals with pathogenic BRCA1 or BRCA2 variants remains relatively unknown. We conducted the largest genome-wide analysis of CNVs in 15,342 BRCA1 and 10,740 BRCA2 pathogenic variant carriers. We used these results to prioritise a candidate breast cancer risk-modifier gene for laboratory analysis and biological validation. Notably, the HR for deletions in BRCA1 suggested an elevated breast cancer risk estimate (hazard ratio (HR) = 1.21), 95% confidence interval (95% CI = 1.09-1.35) compared with non-CNV pathogenic variants. In contrast, deletions overlapping SULT1A1 suggested a decreased breast cancer risk (HR = 0.73, 95% CI 0.59-0.91) in BRCA1 pathogenic variant carriers. Functional analyses of SULT1A1 showed that reduced mRNA expression in pathogenic BRCA1 variant cells was associated with reduced cellular proliferation and reduced DNA damage after treatment with DNA damaging agents. These data provide evidence that deleterious variants in BRCA1 plus SULT1A1 deletions contribute to variable breast cancer risk in BRCA1 carriers.

PMID:36203093 | DOI:10.1038/s42003-022-03978-6

J Exp Psychol Appl. 2022 Oct 6. doi: 10.1037/xap0000450. Online ahead of print.

ABSTRACT

Although theories of emotion associate negative emotional symptoms with cognitive biases in information processing, they rarely specify the details. Here, we characterize cognitive biases in information processing of pleasant and unpleasant information, and how these biases covary with anxious and depressive symptoms, while controlling for general stress and cognitive ability. Forty undergraduates provided emotional symptom scores (Depression Anxiety Stress Scale-21) and performed a statistical learning task that required predicting the next sound in a long sequence of either pleasant or unpleasant naturalistic sounds (blocks). We used an information weights framework to determine if the degree of behavioral change associated with observing either confirmatory (“B” follows “A”) or disconfirmatory (“B” does not follow “A”) transitions differs for pleasant and unpleasant sounds. Bayesian mixed-effects models revealed that negative emotional symptom scores predicted performance as well as processing biases of pleasant and unpleasant information. Further, information weights differed between pleasant and unpleasant information, and importantly, this difference varied based on symptom scores. For example, higher depressive symptom scores predicted a bias of underutilizing disconfirmatory information in unpleasant content. These findings have implications for models of emotional disorders by offering a mechanistic explanation and formalization of the associated cognitive biases. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:36201842 | DOI:10.1037/xap0000450

J Proteome Res. 2022 Oct 6. doi: 10.1021/acs.jproteome.2c00278. Online ahead of print.

ABSTRACT

Combining robust proteomics instrumentation with high-throughput enabling liquid chromatography (LC) systems (e.g., timsTOF Pro and the Evosep One system, respectively) enabled mapping the proteomes of 1000s of samples. Fragpipe is one of the few computational protein identification and quantification frameworks that allows for the time-efficient analysis of such large data sets. However, it requires large amounts of computational power and data storage space that leave even state-of-the-art workstations underpowered when it comes to the analysis of proteomics data sets with 1000s of LC mass spectrometry runs. To address this issue, we developed and optimized a Fragpipe-based analysis strategy for a high-performance computing environment and analyzed 3348 plasma samples (6.4 TB) that were longitudinally collected from hospitalized COVID-19 patients under the auspice of the Immunophenotyping Assessment in a COVID-19 Cohort (IMPACC) study. Our parallelization strategy reduced the total runtime by ∼90% from 116 (theoretical) days to just 9 days in the high-performance computing environment. All code is open-source and can be deployed in any Simple Linux Utility for Resource Management (SLURM) high-performance computing environment, enabling the analysis of large-scale high-throughput proteomics studies.

PMID:36201825 | DOI:10.1021/acs.jproteome.2c00278

J Pers Soc Psychol. 2022 Nov;123(5):1089-1137. doi: 10.1037/pspp0000437.

ABSTRACT

Attachment theory provides a framework for understanding the correlations among interpersonal relationships, stress, and health. Moreover, adult attachment is an important predictor of mental health. However, there is a lack of systematic reviews that simultaneously examine the associations between adult attachment and both positive and negative indicators of mental health. Consequently, we meta-analyzed 224 studies examining the associations between adult attachment and mental health, using robust variance estimation with random effects. The results (k = 245 samples, N = 79,722) showed that higher levels of attachment anxiety and avoidance were positively correlated with negative affect (e.g., depression, anxiety, loneliness) and they were negatively correlated with positive affect (e.g., life satisfaction, self-esteem). More specifically, there were moderate associations between attachment avoidance and negative mental health (r = .28) and positive mental health (r = -.24). Likewise, there were moderate associations between attachment anxiety and negative mental health (r = .42) and positive mental health (r = -.29). Furthermore, the association between the attachment dimensions and mental health outcomes was also moderated by several variables (e.g., gender, age). Finally, these associations remained statistically significant even when the attachment dimensions were mutually controlled using meta-analytic structural equation modeling. Overall, attachment anxiety had larger associations with mental health than did attachment avoidance. Thus, the current results support robust links between adult attachment and mental health. This may have implications for future research and mental health treatments. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:36201836 | DOI:10.1037/pspp0000437

Psychol Methods. 2022 Oct 6. doi: 10.1037/met0000524. Online ahead of print.

ABSTRACT

Structural equation modeling is one of the most popular statistical frameworks in the social and behavioral sciences. Often, detection of groups with distinct sets of parameters in structural equation models (SEM) are of key importance for applied researchers, for example, when investigating differential item functioning for a mental ability test or examining children with exceptional educational trajectories. In the present article, we present a new approach combining subgroup discovery-a well-established toolkit of supervised learning algorithms and techniques from the field of computer science-with structural equation models termed SubgroupSEM. We provide an overview and comparison of three approaches to modeling and detecting heterogeneous groups in structural equation models, namely, finite mixture models, SEM trees, and SubgroupSEM. We provide a step-by-step guide to applying subgroup discovery techniques for structural equation models, followed by a detailed and illustrated presentation of pruning strategies and four subgroup discovery algorithms. Finally, the SubgroupSEM approach will be illustrated on two real data examples, examining measurement invariance of a mental ability test and investigating interesting subgroups for the mediated relationship between predictors of educational outcomes and the trajectories of math competencies in 5th grade children. The illustrative examples are accompanied by examples of the R package subgroupsem, which is a viable implementation of our approach for applied researchers. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:36201823 | DOI:10.1037/met0000524

Psychol Methods. 2022 Oct 6. doi: 10.1037/met0000529. Online ahead of print.

ABSTRACT

Model selection is an omnipresent issue in structural equation modeling (SEM). When deciding among competing theories instantiated as formal statistical models, a trade-off is often sought between goodness-of-fit and model parsimony. Whereas traditional fit assessment in SEM quantifies parsimony solely as the number of free parameters, the ability of a model to account for diverse data patterns-known as fitting propensity-also depends on the functional form of a model. The present investigation provides a systematic assessment of the fitting propensity of models typically considered and compared in SEM, namely, exploratory and confirmatory factor analysis models positing a different number of latent factors or a different hierarchical structure (single-factor, correlated factors, higher-order, and bifactor models). Furthermore, the behavior of commonly used fit indices (CFI, SRMR, RMSEA, TLI) and information criteria (AIC, BIC) in accounting for fitting propensity was assessed. Although the results demonstrated varying degrees of fitting propensity for the models under scrutiny, these differences were mostly driven by the number of free parameters. There was little evidence for additional differences in the functional form of the compared models. Fit indices adjusting for the number of free parameters such as the RMSEA and TLI thus adequately accounted for differences in fitting propensity. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:36201821 | DOI:10.1037/met0000529

Psychol Methods. 2022 Oct 6. doi: 10.1037/met0000532. Online ahead of print.

ABSTRACT

Studies of interaction effects are of great interest because they identify crucial interplay between predictors in explaining outcomes. Previous work has considered several potential sources of statistical bias and substantive misinterpretation in the study of interactions, but less attention has been devoted to the role of the outcome variable in such research. Here, we consider bias and false discovery associated with estimates of interaction parameters as a function of the distributional and metric properties of the outcome variable. We begin by illustrating that, for a variety of noncontinuously distributed outcomes (i.e., binary and count outcomes), attempts to use the linear model for recovery leads to catastrophic levels of bias and false discovery. Next, focusing on transformations of normally distributed variables (i.e., censoring and noninterval scaling), we show that linear models again produce spurious interaction effects. We provide explanations offering geometric and algebraic intuition as to why interactions are a challenge for these incorrectly specified models. In light of these findings, we make two specific recommendations. First, a careful consideration of the outcome’s distributional properties should be a standard component of interaction studies. Second, researchers should approach research focusing on interactions with heightened levels of scrutiny. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:36201820 | DOI:10.1037/met0000532

Obstet Gynecol. 2022 Oct 6. doi: 10.1097/AOG.0000000000004939. Online ahead of print.

ABSTRACT

In this retrospective cohort study, we investigated the relationship between delayed presentation of first-trimester ultrasonographic landmarks of intrauterine pregnancy and perinatal outcomes. Patients presenting as pregnancies of unknown location who ultimately had intrauterine pregnancies were included and divided into two groups, determined by visualization of intrauterine landmarks at hCG <2000 or ≥ 2000. From 487 total patients, there was no significant difference in incidence of favorable perinatal outcome (73.3% vs 73.7%, RR=1.01, 95% CI 0.98-1.10). Of 439 live births, mean birthweight was statistically significantly lower by 115 g in the latter group. No significant difference was found for other neonatal or maternal outcomes. Our findings suggest no relationship between delayed presentation of intrauterine landmarks and poor perinatal outcomes, but a potential association with lower birthweight, though this may have limited clinical significance.

PMID:36201786 | DOI:10.1097/AOG.0000000000004939

Obstet Gynecol. 2022 Oct 6. doi: 10.1097/AOG.0000000000004965. Online ahead of print.

ABSTRACT

OBJECTIVE: To identify rates of fetal autopsy in the United States as well as demographic and clinical characteristics related to consent to autopsy after stillbirth.

METHODS: This is a population-based retrospective cohort study using U.S. fetal death certificates for stillborn fetuses (20 weeks of gestation or more) delivered between January 2014 and December 2016. Multiple gestations were excluded. Fetal autopsy rates were calculated by gestational age, maternal age, self-reported race and ethnicity, education, and having at least one living child. Multivariate logistic regression to adjust for potential confounders was performed.

RESULTS: There were 60,328 stillbirths meeting inclusion criteria. Overall, fetal autopsy was performed in 20.9% of stillbirths. Non-Hispanic Black women had the highest rate of fetal autopsy (22.9%, 95% CI 22.3-23.6%), compared with non-Hispanic White women (20.4%, 95% CI 20.0-20.9%) and Hispanic women (19.6%, 95% CI 19.0-20.3%) (P<.001). After adjusting for potential confounders, maternal non-Hispanic Black race (adjusted odds ratio [aOR] 1.22, 95% CI 1.16-1.29), higher education (graduate degree: aOR 1.62, 95% CI 1.47-1.79), and higher gestational age (term: aOR 2.08, 95% CI 1.95-2.23) were associated with increased aORs for fetal autopsy. Maternal age 40 years or older (aOR 0.77 95% CI 0.63-0.92) and having at least one living child (aOR 0.74, 95% CI 0.71-0.78) were associated with a decreased aOR of having a fetal autopsy. Women of American Indian or Alaska Native decent had decreased uptake of fetal autopsy compared with non-Hispanic White women (aOR 0.72, 95% CI 0.58-0.90).

CONCLUSION: Fetal autopsy rates are low throughout the United States. The reasons for low autopsy rates warrant further exploration to inform strategies to increase availability and uptake.

PMID:36201780 | DOI:10.1097/AOG.0000000000004965

Obstet Gynecol. 2022 Oct 6. doi: 10.1097/AOG.0000000000004958. Online ahead of print.

ABSTRACT

OBJECTIVE: To describe the testing rate, patient characteristics, temporal trends, timing, and results of germline and somatic BRCA testing in patients with ovarian cancer using real-world data.

METHODS: We included a cross-sectional subset of adult patients diagnosed with ovarian cancer between January 1, 2011, and November 30, 2018, who received frontline treatment and were followed for at least 1 year in a real-world database. The primary outcome was receipt of BRCA testing, classified by biosample source as germline (blood or saliva) or somatic (tissue). Lines of therapy (frontline, second line, third line) were derived based on dates of surgery and chemotherapy. Descriptive statistics were analyzed.

RESULTS: Among 2,557 patients, 72.2% (n=1,846) had at least one documented BRCA test. Among tested patients, 62.5% (n=1,154) had only germline testing, 10.6% (n=197) had only somatic testing, and 19.9% (n=368) had both. Most patients had testing before (9.7%, n=276) or during (48.6%, n=1,521) frontline therapy, with 17.6% (n=273) tested during second-line and 12.7% (n=129) tested during third-line therapy. Patients who received BRCA testing, compared with patients without testing, were younger (mean age 63 years vs 66 years, P<.001) and were more likely to be treated at an academic practice (10.4% vs 7.0%, P=.01), with differences by Eastern Cooperative Oncology Group performance score (P<.001), stage of disease (P<.001), histology (P<.001), geography (P<.001), and type of frontline therapy (P<.001), but no differences based on race or ethnicity. The proportion of patients who received BRCA testing within 1 year of diagnosis increased from 24.6% of patients in 2011 to 75.6% of patients in 2018.

CONCLUSION: In a large cohort of patients with ovarian cancer, significant practice disparities existed in testing for actionable BRCA mutations. Despite increased testing over time, many patients did not receive testing, suggesting missed opportunities to identify patients appropriate for targeted therapy and genetic counseling.

PMID:36201776 | DOI:10.1097/AOG.0000000000004958