An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics.

For a decade, The Cancer Genome Atlas (TCGA) program collected clinicopathologic annotation data along with multi-platform molecular profiles of more than 11,000 human tumors across 33 different cancer types. TCGA clinical data contain key features representing the democratized nature of the data collection process. To ensure proper use of this large clinical dataset associated with genomic features, we developed a standardized dataset named the TCGA Pan-Cancer Clinical Data Resource (TCGA-CDR), which includes four major clinical outcome endpoints. In addition to detailing major challenges and statistical limitations encountered during the effort of integrating the acquired clinical data, we present a summary that includes endpoint usage recommendations for each cancer type. These TCGA-CDR findings appear to be consistent with cancer genomics studies independent of the TCGA effort and provide opportunities for investigating cancer biology using clinical correlates at an unprecedented scale.

Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation.

Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation.

Multiplatform analysis of 12 cancer types reveals molecular classification within and across tissues of origin.

Recent genomic analyses of pathologically defined tumor types identify "within-a-tissue" disease subtypes. However, the extent to which genomic signatures are shared across tissues is still unclear. We performed an integrative analysis using five genome-wide platforms and one proteomic platform on 3,527 specimens from 12 cancer types, revealing a unified classification into 11 major subtypes. Five subtypes were nearly identical to their tissue-of-origin counterparts, but several distinct cancer types were found to converge into common subtypes. Lung squamous, head and neck, and a subset of bladder cancers coalesced into one subtype typified by TP53 alterations, TP63 amplifications, and high expression of immune and proliferation pathway genes. Of note, bladder cancers split into three pan-cancer subtypes. The multiplatform classification, while correlated with tissue-of-origin, provides independent information for predicting clinical outcomes. All data sets are available for data-mining from a unified resource to support further biological discoveries and insights into novel therapeutic strategies.

Clinical Validation of Digitally Acquired Clinical Data and Machine Learning Models for Remote Measurement of Psoriasis and Psoriatic Arthritis: A Proof-of-Concept Study.

OBJECTIVE: Psoriatic disease remains underdiagnosed and undertreated. We developed and validated a suite of novel, sensor-based smartphone assessments (Psorcast app) that can be self-administered to measure cutaneous and musculoskeletal signs and symptoms of psoriatic disease. METHODS: Participants with psoriasis (PsO) or psoriatic arthritis (PsA) and healthy controls were recruited between June 5, 2019, and November 10, 2021, at 2 academic medical centers. Concordance and accuracy of digital measures and image-based machine learning models were compared to their analogous clinical measures from trained rheumatologists and dermatologists. RESULTS: Of 104 study participants, 51 (49%) were female and 53 (51%) were male, with a mean age of 42.3 years (SD 12.6). Seventy-nine (76%) participants had PsA, 16 (15.4%) had PsO, and 9 (8.7%) were healthy controls. Digital patient assessment of percent body surface area (BSA) affected with PsO demonstrated very strong concordance (Lin concordance correlation coefficient [CCC] 0.94 [95% CI 0.91-0.96]) with physician-assessed BSA. The in-clinic and remote target plaque physician global assessments showed fair-to-moderate concordance (CCCerythema 0.72 [0.59-0.85]; CCCinduration 0.72 [0.62-0.82]; CCCscaling 0.60 [0.48-0.72]). Machine learning models of hand photos taken by patients accurately identified clinically diagnosed nail PsO with an accuracy of 0.76. The Digital Jar Open assessment categorized physician-assessed upper extremity involvement, considering joint tenderness or enthesitis (AUROC 0.68 [0.47-0.85]). CONCLUSION: The Psorcast digital assessments achieved significant clinical validity, although they require further validation in larger cohorts before use in evidence-based medicine or clinical trial settings. The smartphone software and analysis pipelines from the Psorcast suite are open source and freely available.

Mobile toolbox (MTB) remote measures of executive function and processing speed: development and validation.

OBJECTIVE: The ability to remotely monitor cognitive skills is increasing with the ubiquity of smartphones. The Mobile Toolbox (MTB) is a new measurement system that includes measures assessing Executive Functioning (EF) and Processing Speed (PS): Arrow Matching, Shape-Color Sorting, and Number-Symbol Match. The purpose of this study was to assess their psychometric properties. METHOD: MTB measures were developed for smartphone administration based on constructs measured in the NIH Toolbox® (NIHTB). Psychometric properties of the resulting measures were evaluated in three studies with participants ages 18 to 90. In Study 1 (N = 92), participants completed MTB measures in the lab and were administered both equivalent NIH TB measures and other external measures of similar cognitive constructs. In Study 2 (N = 1,021), participants completed the equivalent NIHTB measures in the lab and then took the MTB measures on their own, remotely. In Study 3 (N = 168), participants completed MTB measures twice remotely, two weeks apart. RESULTS: All three measures exhibited very high internal consistency and strong test-retest reliability, as well as moderately high correlations with comparable NIHTB tests and moderate correlations with external measures of similar constructs. Phone operating system (iOS vs. Android) had a significant impact on performance for Arrow Matching and Shape-Color Sorting, but no impact on either validity or reliability. CONCLUSIONS: Results support the reliability and convergent validity of MTB EF and PS measures for use across the adult lifespan in remote, self-administered designs.

A novel estimator for the two-way partial AUC.

BACKGROUND: The two-way partial AUC has been recently proposed as a way to directly quantify partial area under the ROC curve with simultaneous restrictions on the sensitivity and specificity ranges of diagnostic tests or classifiers. The metric, as originally implemented in the tpAUC R package, is estimated using a nonparametric estimator based on a trimmed Mann-Whitney U-statistic, which becomes computationally expensive in large sample sizes. (Its computational complexity is of order [Formula: see text], where [Formula: see text] and [Formula: see text] represent the number of positive and negative cases, respectively). This is problematic since the statistical methodology for comparing estimates generated from alternative diagnostic tests/classifiers relies on bootstrapping resampling and requires repeated computations of the estimator on a large number of bootstrap samples. METHODS: By leveraging the graphical and probabilistic representations of the AUC, partial AUCs, and two-way partial AUC, we derive a novel estimator for the two-way partial AUC, which can be directly computed from the output of any software able to compute AUC and partial AUCs. We implemented our estimator using the computationally efficient pROC R package, which leverages a nonparametric approach using the trapezoidal rule for the computation of AUC and partial AUC scores. (Its computational complexity is of order [Formula: see text], where [Formula: see text].). We compare the empirical bias and computation time of the proposed estimator against the original estimator provided in the tpAUC package in a series of simulation studies and on two real datasets. RESULTS: Our estimator tended to be less biased than the original estimator based on the trimmed Mann-Whitney U-statistic across all experiments (and showed considerably less bias in the experiments based on small sample sizes). But, most importantly, because the computational complexity of the proposed estimator is of order [Formula: see text], rather than [Formula: see text], it is much faster to compute when sample sizes are large. CONCLUSIONS: The proposed estimator provides an improvement for the computation of two-way partial AUC, and allows the comparison of diagnostic tests/machine learning classifiers in large datasets where repeated computations of the original estimator on bootstrap samples become too expensive to compute.

The STEPWISE study: study protocol for a smartphone-based exercise solution for people with Parkinson's Disease (randomized controlled trial).

BACKGROUND: Exercise has various health benefits for people with Parkinson's disease (PD). However, implementing exercise into daily life and long-term adherence remain challenging. To increase a sustainable engagement with physical activity of people with PD, interventions that are motivating, accessible, and scalable are needed. We primarily aim to investigate whether a smartphone app (STEPWISE app) can increase physical activity (i.e., step count) in people with PD over one year. Our second aim is to investigate the potential effects of the intervention on physical fitness, and motor- and non-motor function. Our third aim is to explore whether there is a dose-response relationship between volume of physical activity and our secondary endpoints. METHODS: STEPWISE is a double-blind, randomized controlled trial. We aim to include 452 Dutch people with PD who can walk independently (Hoehn & Yahr stages 1-3) and who do not take more than 7,000 steps per day prior to inclusion. Physical activity levels are measured as step counts on the participant's own smartphone and scaled as percentage of each participant's baseline. Participants are randomly assigned to an active control group with an increase of 5-20% (active controls) or any of the three intervention arms with increases of 25-100% (intermediate dose), 50-200% (large dose), or 100-400% (very large dose). The primary endpoint is change in step count as measured by the STEPWISE smartphone app from baseline to 52 weeks. For our primary aim, we will evaluate the between-group difference in average daily step count change from baseline to 52 weeks. For our second aim, measures of physical fitness, and motor- and non-motor function are included. For our third aim, we will associate 52-week changes in step count with 52-week changes in secondary outcomes. DISCUSSION: This trial evaluates the potential of a smartphone-based intervention to increase activity levels in people with PD. We envision that motivational apps will increase adherence to physical activity recommendations and could permit conduct of remote clinical trials of exercise for people with PD or those at risk of PD. TRIAL REGISTRATION: ClinicalTrials.gov; NCT04848077; 19/04/2021. CLINICALTRIALS: gov/ct2/show/NCT04848077.

Identifying and ranking potential driver genes of Alzheimer's disease using multiview evidence aggregation.

MOTIVATION: Late onset Alzheimer's disease is currently a disease with no known effective treatment options. To better understand disease, new multi-omic data-sets have recently been generated with the goal of identifying molecular causes of disease. However, most analytic studies using these datasets focus on uni-modal analysis of the data. Here, we propose a data driven approach to integrate multiple data types and analytic outcomes to aggregate evidences to support the hypothesis that a gene is a genetic driver of the disease. The main algorithmic contributions of our article are: (i) a general machine learning framework to learn the key characteristics of a few known driver genes from multiple feature sets and identifying other potential driver genes which have similar feature representations, and (ii) A flexible ranking scheme with the ability to integrate external validation in the form of Genome Wide Association Study summary statistics. While we currently focus on demonstrating the effectiveness of the approach using different analytic outcomes from RNA-Seq studies, this method is easily generalizable to other data modalities and analysis types. RESULTS: We demonstrate the utility of our machine learning algorithm on two benchmark multiview datasets by significantly outperforming the baseline approaches in predicting missing labels. We then use the algorithm to predict and rank potential drivers of Alzheimer's. We show that our ranked genes show a significant enrichment for single nucleotide polymorphisms associated with Alzheimer's and are enriched in pathways that have been previously associated with the disease. AVAILABILITY AND IMPLEMENTATION: Source code and link to all feature sets is available at https://github.com/Sage-Bionetworks/EvidenceAggregatedDriverRanking.

Unsupervised Analysis of Transcriptomics in Bacterial Sepsis Across Multiple Datasets Reveals Three Robust Clusters.

OBJECTIVES: To find and validate generalizable sepsis subtypes using data-driven clustering. DESIGN: We used advanced informatics techniques to pool data from 14 bacterial sepsis transcriptomic datasets from eight different countries (n = 700). SETTING: Retrospective analysis. SUBJECTS: Persons admitted to the hospital with bacterial sepsis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A unified clustering analysis across 14 discovery datasets revealed three subtypes, which, based on functional analysis, we termed "Inflammopathic, Adaptive, and Coagulopathic." We then validated these subtypes in nine independent datasets from five different countries (n = 600). In both discovery and validation data, the Adaptive subtype is associated with a lower clinical severity and lower mortality rate, and the Coagulopathic subtype is associated with higher mortality and clinical coagulopathy. Further, these clusters are statistically associated with clusters derived by others in independent single sepsis cohorts. CONCLUSIONS: The three sepsis subtypes may represent a unifying framework for understanding the molecular heterogeneity of the sepsis syndrome. Further study could potentially enable a precision medicine approach of matching novel immunomodulatory therapies with septic patients most likely to benefit.

Population genetic inference from personal genome data: impact of ancestry and admixture on human genomic variation.

Full sequencing of individual human genomes has greatly expanded our understanding of human genetic variation and population history. Here, we present a systematic analysis of 50 human genomes from 11 diverse global populations sequenced at high coverage. Our sample includes 12 individuals who have admixed ancestry and who have varying degrees of recent (within the last 500 years) African, Native American, and European ancestry. We found over 21 million single-nucleotide variants that contribute to a 1.75-fold range in nucleotide heterozygosity across diverse human genomes. This heterozygosity ranged from a high of one heterozygous site per kilobase in west African genomes to a low of 0.57 heterozygous sites per kilobase in segments inferred to have diploid Native American ancestry from the genomes of Mexican and Puerto Rican individuals. We show evidence of all three continental ancestries in the genomes of Mexican, Puerto Rican, and African American populations, and the genome-wide statistics are highly consistent across individuals from a population once ancestry proportions have been accounted for. Using a generalized linear model, we identified subtle variations across populations in the proportion of neutral versus deleterious variation and found that genome-wide statistics vary in admixed populations even once ancestry proportions have been factored in. We further infer that multiple periods of gene flow shaped the diversity of admixed populations in the Americas-70% of the European ancestry in today's African Americans dates back to European gene flow happening only 7-8 generations ago.

Patterns of ancestry, signatures of natural selection, and genetic association with stature in Western African pygmies.

African Pygmy groups show a distinctive pattern of phenotypic variation, including short stature, which is thought to reflect past adaptation to a tropical environment. Here, we analyze Illumina 1M SNP array data in three Western Pygmy populations from Cameroon and three neighboring Bantu-speaking agricultural populations with whom they have admixed. We infer genome-wide ancestry, scan for signals of positive selection, and perform targeted genetic association with measured height variation. We identify multiple regions throughout the genome that may have played a role in adaptive evolution, many of which contain loci with roles in growth hormone, insulin, and insulin-like growth factor signaling pathways, as well as immunity and neuroendocrine signaling involved in reproduction and metabolism. The most striking results are found on chromosome 3, which harbors a cluster of selection and association signals between approximately 45 and 60 Mb. This region also includes the positional candidate genes DOCK3, which is known to be associated with height variation in Europeans, and CISH, a negative regulator of cytokine signaling known to inhibit growth hormone-stimulated STAT5 signaling. Finally, pathway analysis for genes near the strongest signals of association with height indicates enrichment for loci involved in insulin and insulin-like growth factor signaling.

Impacts on study design when implementing digital measures in Parkinson's disease-modifying therapy trials.

Introduction: Parkinson's Disease affects over 8.5 million people and there are currently no medications approved to treat underlying disease. Clinical trials for disease modifying therapies (DMT) are hampered by a lack of sufficiently sensitive measures to detect treatment effect. Reliable digital assessments of motor function allow for frequent at-home measurements that may be able to sensitively detect disease progression. Methods: Here, we estimate the test-retest reliability of a suite of at-home motor measures derived from raw triaxial accelerometry data collected from 44 participants (21 with confirmed PD) and use the estimates to simulate digital measures in DMT trials. We consider three schedules of assessments and fit linear mixed models to the simulated data to determine whether a treatment effect can be detected. Results: We find at-home measures vary in reliability; many have ICCs as high as or higher than MDS-UPDRS part III total score. Compared with quarterly in-clinic assessments, frequent at-home measures reduce the sample size needed to detect a 30% reduction in disease progression from over 300 per study arm to 150 or less than 100 for bursts and evenly spaced at-home assessments, respectively. The results regarding superiority of at-home assessments for detecting change over time are robust to relaxing assumptions regarding the responsiveness to disease progression and variability in progression rates. Discussion: Overall, at-home measures have a favorable reliability profile for sensitive detection of treatment effects in DMT trials. Future work is needed to better understand the causes of variability in PD progression and identify the most appropriate statistical methods for effect detection.

Development and validation of an episodic memory measure in the Mobile Toolbox (MTB): Arranging Pictures.

INTRODUCTION: Arranging Pictures is a new episodic memory test based on the NIH Toolbox (NIHTB) Picture Sequence Memory measure and optimized for self-administration on a personal smartphone within the Mobile Toolbox (MTB). We describe evidence from three distinct validation studies. METHOD: In Study 1, 92 participants self-administered Arranging Pictures on study-provided smartphones in the lab and were administered external measures of similar and dissimilar constructs by trained examiners to assess validity under controlled circumstances. In Study 2, 1,021 participants completed the external measures in the lab and self-administered Arranging Pictures remotely on their personal smartphones to assess validity in real-world contexts. In Study 3, 141 participants self-administered Arranging Pictures remotely twice with a two-week delay on personal iOS smartphones to assess test-retest reliability and practice effects. RESULTS: Internal consistency was good across samples (ρxx = .80 to .85, p < .001). Test-retest reliability was marginal (ICC = .49, p < .001) and there were significant practice effects after a two-week delay (ΔM = 3.21 (95% CI [2.56, 3.88]). As expected, correlations with convergent measures were significant and moderate to large in magnitude (ρ = .44 to .76, p < .001), while correlations with discriminant measures were small (ρ = .23 to .27, p < .05) or nonsignificant. Scores demonstrated significant negative correlations with age (ρ = -.32 to -.21, p < .001). Mean performance was slightly higher in the iOS compared to the Android group (MiOS = 18.80, NiOS = 635; MAndroid = 17.11, NAndroid = 386; t(757.73) = 4.17, p < .001), but device type did not significantly influence the psychometric properties of the measure. Indicators of potential cheating were mixed; average scores were significantly higher in the remote samples (F(2, 850) = 11.415, p < .001), but there were not significantly more perfect scores. CONCLUSION: The MTB Arranging Pictures measure demonstrated evidence of reliability and validity when self-administered on personal device. Future research should examine the potential for cheating in remote settings and the properties of the measure in clinical samples.

Development and Validation of a Vocabulary Measure in the Mobile Toolbox.

OBJECTIVE: We describe the development of a new computer adaptive vocabulary test, Mobile Toolbox (MTB) Word Meaning, and validity evidence from 3 studies. METHOD: Word Meaning was designed to be a multiple-choice synonym test optimized for self-administration on a personal smartphone. The items were first calibrated online in a sample of 7,525 participants to create the computer-adaptive test algorithm for the Word Meaning measure within the MTB app. In Study 1, 92 participants self-administered Word Meaning on study-provided smartphones in the lab and were administered external measures by trained examiners. In Study 2, 1,021 participants completed the external measures in the lab and Word Meaning was self-administered remotely on their personal smartphones. In Study 3, 141 participants self-administered Word Meaning remotely twice with a 2-week delay on personal iPhones. RESULTS: The final bank included 1363 items. Internal consistency was adequate to good across samples (ρxx = 0.78 to 0.81, p < .001). Test-retest reliability was good (ICC = 0.65, p < .001), and the mean theta score was not significantly different upon the second administration. Correlations were moderate to large with measures of similar constructs (ρ = 0.67-0.75, p < .001) and non-significant with measures of dissimilar constructs. Scores demonstrated small to moderate correlations with age (ρ = 0.35 to 0.45, p < .001) and education (ρ = 0.26, p < .001). CONCLUSION: The MTB Word Meaning measure demonstrated evidence of reliability and validity in three samples. Further validation studies in clinical samples are necessary.

Developing better digital health measures of Parkinson's disease using free living data and a crowdsourced data analysis challenge.

One of the promising opportunities of digital health is its potential to lead to more holistic understandings of diseases by interacting with the daily life of patients and through the collection of large amounts of real-world data. Validating and benchmarking indicators of disease severity in the home setting is difficult, however, given the large number of confounders present in the real world and the challenges in collecting ground truth data in the home. Here we leverage two datasets collected from patients with Parkinson's disease, which couples continuous wrist-worn accelerometer data with frequent symptom reports in the home setting, to develop digital biomarkers of symptom severity. Using these data, we performed a public benchmarking challenge in which participants were asked to build measures of severity across 3 symptoms (on/off medication, dyskinesia, and tremor). 42 teams participated and performance was improved over baseline models for each subchallenge. Additional ensemble modeling across submissions further improved performance, and the top models validated in a subset of patients whose symptoms were observed and rated by trained clinicians.

RNA-Seq quantification of the human small airway epithelium transcriptome.

BACKGROUND: The small airway epithelium (SAE), the cell population that covers the human airway surface from the 6th generation of airway branching to the alveoli, is the major site of lung disease caused by smoking. The focus of this study is to provide quantitative assessment of the SAE transcriptome in the resting state and in response to chronic cigarette smoking using massive parallel mRNA sequencing (RNA-Seq). RESULTS: The data demonstrate that 48% of SAE expressed genes are ubiquitous, shared with many tissues, with 52% enriched in this cell population. The most highly expressed gene, SCGB1A1, is characteristic of Clara cells, the cell type unique to the human SAE. Among other genes expressed by the SAE are those related to Clara cell differentiation, secretory mucosal defense, and mucociliary differentiation. The high sensitivity of RNA-Seq permitted quantification of gene expression related to infrequent cell populations such as neuroendocrine cells and epithelial stem/progenitor cells. Quantification of the absolute smoking-induced changes in SAE gene expression revealed that, compared to ubiquitous genes, more SAE-enriched genes responded to smoking with up-regulation, and those with the highest basal expression levels showed most dramatic changes. Smoking had no effect on SAE gene splicing, but was associated with a shift in molecular pattern from Clara cell-associated towards the mucus-secreting cell differentiation pathway with multiple features of cancer-associated molecular phenotype. CONCLUSIONS: These observations provide insights into the unique biology of human SAE by providing quantitative assessment of the global transcriptome under physiological conditions and in response to the stress of chronic cigarette smoking.

Inferring genome-wide patterns of admixture in Qataris using fifty-five ancestral populations.

BACKGROUND: Populations of the Arabian Peninsula have a complex genetic structure that reflects waves of migrations including the earliest human migrations from Africa and eastern Asia, migrations along ancient civilization trading routes and colonization history of recent centuries. RESULTS: Here, we present a study of genome-wide admixture in this region, using 156 genotyped individuals from Qatar, a country located at the crossroads of these migration patterns. Since haplotypes of these individuals could have originated from many different populations across the world, we have developed a machine learning method "SupportMix" to infer loci-specific genomic ancestry when simultaneously analyzing many possible ancestral populations. Simulations show that SupportMix is not only more accurate than other popular admixture discovery tools but is the first admixture inference method that can efficiently scale for simultaneous analysis of 50-100 putative ancestral populations while being independent of prior demographic information. CONCLUSIONS: By simultaneously using the 55 world populations from the Human Genome Diversity Panel, SupportMix was able to extract the fine-scale ancestry of the Qatar population, providing many new observations concerning the ancestry of the region. For example, as well as recapitulating the three major sub-populations in Qatar, composed of mainly Arabic, Persian, and African ancestry, SupportMix additionally identifies the specific ancestry of the Persian group to populations sampled in Greater Persia rather than from China and the ancestry of the African group to sub-Saharan origin and not Southern African Bantu origin as previously thought.

PCAdmix: principal components-based assignment of ancestry along each chromosome in individuals with admixed ancestry from two or more populations.

Identifying ancestry along each chromosome in admixed individuals provides a wealth of information for understanding the population genetic history of admixture events and is valuable for admixture mapping and identifying recent targets of selection. We present PCAdmix (available at https://sites.google.com/site/pcadmix/home ), a Principal Components-based algorithm for determining ancestry along each chromosome from a high-density, genome-wide set of phased single-nucleotide polymorphism (SNP) genotypes of admixed individuals. We compare our method to HAPMIX on simulated data from two ancestral populations, and we find high concordance between the methods. Our method also has better accuracy than LAMP when applied to three-population admixture, a situation as yet unaddressed by HAPMIX. Finally, we apply our method to a data set of four Latino populations with European, African, and Native American ancestry. We find evidence of assortative mating in each of the four populations, and we identify regions of shared ancestry that may be recent targets of selection and could serve as candidate regions for admixture-based association mapping.

Disentangling personalized treatment effects from "time-of-the-day" confounding in mobile health studies.

Ideally, a patient's response to medication can be monitored by measuring changes in performance of some activity. In observational studies, however, any detected association between treatment ("on-medication" vs "off-medication") and the outcome (performance in the activity) might be due to confounders. In particular, causal inferences at the personalized level are especially vulnerable to confounding effects that arise in a cyclic fashion. For quick acting medications, effects can be confounded by circadian rhythms and daily routines. Using the time-of-the-day as a surrogate for these confounders and the performance measurements as captured on a smartphone, we propose a personalized statistical approach to disentangle putative treatment and "time-of-the-day" effects, that leverages conditional independence relations spanned by causal graphical models involving the treatment, time-of-the-day, and outcome variables. Our approach is based on conditional independence tests implemented via standard and temporal linear regression models. Using synthetic data, we investigate when and how residual autocorrelation can affect the standard tests, and how time series modeling (namely, ARIMA and robust regression via HAC covariance matrix estimators) can remedy these issues. In particular, our simulations illustrate that when patients perform their activities in a paired fashion, positive autocorrelation can lead to conservative results for the standard regression approach (i.e., lead to deflated true positive detection), whereas negative autocorrelation can lead to anticonservative behavior (i.e., lead to inflated false positive detection). The adoption of time series methods, on the other hand, leads to well controlled type I error rates. We illustrate the application of our methodology with data from a Parkinson's disease mobile health study.

A Multifaceted benchmarking of synthetic electronic health record generation models.

Synthetic health data have the potential to mitigate privacy concerns in supporting biomedical research and healthcare applications. Modern approaches for data generation continue to evolve and demonstrate remarkable potential. Yet there is a lack of a systematic assessment framework to benchmark methods as they emerge and determine which methods are most appropriate for which use cases. In this work, we introduce a systematic benchmarking framework to appraise key characteristics with respect to utility and privacy metrics. We apply the framework to evaluate synthetic data generation methods for electronic health records data from two large academic medical centers with respect to several use cases. The results illustrate that there is a utility-privacy tradeoff for sharing synthetic health data and further indicate that no method is unequivocally the best on all criteria in each use case, which makes it evident why synthetic data generation methods need to be assessed in context.