Interpretable (not just posthoc-explainable) medical claims modeling for discharge placement to reduce preventable all-cause readmissions or death.

We developed an inherently interpretable multilevel Bayesian framework for representing variation in regression coefficients that mimics the piecewise linearity of ReLU-activated deep neural networks. We used the framework to formulate a survival model for using medical claims to predict hospital readmission and death that focuses on discharge placement, adjusting for confounding in estimating causal local average treatment effects. We trained the model on a 5% sample of Medicare beneficiaries from 2008 and 2011, based on their 2009-2011 inpatient episodes (approximately 1.2 million), and then tested the model on 2012 episodes (approximately 400 thousand). The model scored an out-of-sample AUROC of approximately 0.75 on predicting all-cause readmissions-defined using official Centers for Medicare and Medicaid Services (CMS) methodology-or death within 30-days of discharge, being competitive against XGBoost and a Bayesian deep neural network, demonstrating that one need-not sacrifice interpretability for accuracy. Crucially, as a regression model, it provides what blackboxes cannot-its exact gold-standard global interpretation, explicitly defining how the model performs its internal "reasoning" for mapping the input data features to predictions. In doing so, we identify relative risk factors and quantify the effect of discharge placement. We also show that the posthoc explainer SHAP provides explanations that are inconsistent with the ground truth model reasoning that our model readily admits.

Central disorders of hypersomnolence: diagnostic discrepancies between military and civilian sleep centers.

STUDY OBJECTIVES: The majority of active-duty service members obtain insufficient sleep, which can influence diagnostic evaluations for sleep disorders, including disorders of hypersomnolence. An incorrect diagnosis of hypersomnia may be career ending for military service or lead to inappropriate medical care. This study was conducted to assess the rates at which narcolepsy (Nc) and idiopathic hypersomnia (IH) are diagnosed by military vs civilian sleep disorders centers. METHODS: This retrospective study utilized claims data from the Military Health System Data Repository. The analyses compared diagnostic rates of military personnel by provider type-either civilian provider or military provider-from January 1, 2016 to December 31, 2019. Three diagnostic categories for Nc and IH: Nc or IH, Nc only, and IH only, were assessed with multivariate logistic regression models. RESULTS: We found that among service members evaluated for a sleep disorder, the odds ratios of a positive diagnosis at a civilian facility vs a military facility for Nc or IH was 2.1, for Nc only was 2.1, and IH only was 2.0 over the 4-year period. CONCLUSIONS: Civilian sleep specialists were twice as likely to diagnose central disorders of hypersomnolence compared to military specialists. Raising awareness about this discrepancy is critical given the occupational and patient care-related implications of misdiagnoses. CITATION: Thomas CL, Vattikuti S, Shaha D, et al. Central disorders of hypersomnolence: diagnostic discrepancies between military and civilian sleep centers. J Clin Sleep Med. 2022;18(10):2433-2441.

Global prediction of unreported SARS-CoV2 infection from observed COVID-19 cases.

Estimation of infectiousness and fatality of the SARS-CoV-2 virus in the COVID-19 global pandemic is complicated by ascertainment bias resulting from incomplete and non-representative samples of infected individuals. We developed a strategy for overcoming this bias to obtain more plausible estimates of the true values of key epidemiological variables. We fit mechanistic Bayesian latent-variable SIR models to confirmed COVID-19 cases, deaths, and recoveries, for all regions (countries and US states) independently. Bayesian averaging over models, we find that the raw infection incidence rate underestimates the true rate by a factor, the case ascertainment ratio CARt that depends upon region and time. At the regional onset of COVID-19, the predicted global median was 13 infections unreported for each case confirmed (CARt = 0.07 C.I. (0.02, 0.4)). As the infection spread, the median CARt rose to 9 unreported cases for every one diagnosed as of April 15, 2020 (CARt = 0.1 C.I. (0.02, 0.5)). We also estimate that the median global initial reproduction number R0 is 3.3 (C.I (1.5, 8.3)) and the total infection fatality rate near the onset is 0.17% (C.I. (0.05%, 0.9%)). However the time-dependent reproduction number Rt and infection fatality rate as of April 15 were 1.2 (C.I. (0.6, 2.5)) and 0.8% (C.I. (0.2%,4%)), respectively. We find that there is great variability between country- and state-level values. Our estimates are consistent with recent serological estimates of cumulative infections for the state of New York, but inconsistent with claims that very large fractions of the population have already been infected in most other regions. For most regions, our estimates imply a great deal of uncertainty about the current state and trajectory of the epidemic.

Dynamical modeling of multi-scale variability in neuronal competition.

Variability is observed at multiple-scales in the brain and ubiquitous in perception. However, the nature of perceptual variability is an open question. We focus on variability during perceptual rivalry, a form of neuronal competition. Rivalry provides a window into neural processing since activity in many brain areas is correlated to the alternating perception rather than a constant ambiguous stimulus. It exhibits robust properties at multiple scales including conscious awareness and neuron dynamics. The prevalent theory for spiking variability is called the balanced state; whereas, the source of perceptual variability is unknown. Here we show that a single biophysical circuit model, satisfying certain mutual inhibition architectures, can explain spiking and perceptual variability during rivalry. These models adhere to a broad set of strict experimental constraints at multiple scales. As we show, the models predict how spiking and perceptual variability changes with stimulus conditions.

Canonical Cortical Circuit Model Explains Rivalry, Intermittent Rivalry, and Rivalry Memory.

It has been shown that the same canonical cortical circuit model with mutual inhibition and a fatigue process can explain perceptual rivalry and other neurophysiological responses to a range of static stimuli. However, it has been proposed that this model cannot explain responses to dynamic inputs such as found in intermittent rivalry and rivalry memory, where maintenance of a percept when the stimulus is absent is required. This challenges the universality of the basic canonical cortical circuit. Here, we show that by including an overlooked realistic small nonspecific background neural activity, the same basic model can reproduce intermittent rivalry and rivalry memory without compromising static rivalry and other cortical phenomena. The background activity induces a mutual-inhibition mechanism for short-term memory, which is robust to noise and where fine-tuning of recurrent excitation or inclusion of sub-threshold currents or synaptic facilitation is unnecessary. We prove existence conditions for the mechanism and show that it can explain experimental results from the quartet apparent motion illusion, which is a prototypical intermittent rivalry stimulus.

Uncovering the Genetic Architectures of Quantitative Traits.

The aim of a genome-wide association study (GWAS) is to identify loci in the human genome affecting a phenotype of interest. This review summarizes some recent work on conceptual and methodological aspects of GWAS. The average effect of gene substitution at a given causal site in the genome is the key estimand in GWAS, and we argue for its fundamental importance. Implicit in the definition of average effect is a linear model relating genotype to phenotype. The fraction of the phenotypic variance ascribable to polymorphic sites with nonzero average effects in this linear model is called the heritability, and we describe methods for estimating this quantity from GWAS data. Finally, we show that the theory of compressed sensing can be used to provide a sharp estimate of the sample size required to identify essentially all sites contributing to the heritability of a given phenotype.

Second-generation PLINK: rising to the challenge of larger and richer datasets.

BACKGROUND: PLINK 1 is a widely used open-source C/C++ toolset for genome-wide association studies (GWAS) and research in population genetics. However, the steady accumulation of data from imputation and whole-genome sequencing studies has exposed a strong need for faster and scalable implementations of key functions, such as logistic regression, linkage disequilibrium estimation, and genomic distance evaluation. In addition, GWAS and population-genetic data now frequently contain genotype likelihoods, phase information, and/or multiallelic variants, none of which can be represented by PLINK 1's primary data format. FINDINGS: To address these issues, we are developing a second-generation codebase for PLINK. The first major release from this codebase, PLINK 1.9, introduces extensive use of bit-level parallelism, [Formula: see text]-time/constant-space Hardy-Weinberg equilibrium and Fisher's exact tests, and many other algorithmic improvements. In combination, these changes accelerate most operations by 1-4 orders of magnitude, and allow the program to handle datasets too large to fit in RAM. We have also developed an extension to the data format which adds low-overhead support for genotype likelihoods, phase, multiallelic variants, and reference vs. alternate alleles, which is the basis of our planned second release (PLINK 2.0). CONCLUSIONS: The second-generation versions of PLINK will offer dramatic improvements in performance and compatibility. For the first time, users without access to high-end computing resources can perform several essential analyses of the feature-rich and very large genetic datasets coming into use.

Applying compressed sensing to genome-wide association studies.

BACKGROUND: The aim of a genome-wide association study (GWAS) is to isolate DNA markers for variants affecting phenotypes of interest. This is constrained by the fact that the number of markers often far exceeds the number of samples. Compressed sensing (CS) is a body of theory regarding signal recovery when the number of predictor variables (i.e., genotyped markers) exceeds the sample size. Its applicability to GWAS has not been investigated. RESULTS: Using CS theory, we show that all markers with nonzero coefficients can be identified (selected) using an efficient algorithm, provided that they are sufficiently few in number (sparse) relative to sample size. For heritability equal to one (h (2) = 1), there is a sharp phase transition from poor performance to complete selection as the sample size is increased. For heritability below one, complete selection still occurs, but the transition is smoothed. We find for h (2) ∼ 0.5 that a sample size of approximately thirty times the number of markers with nonzero coefficients is sufficient for full selection. This boundary is only weakly dependent on the number of genotyped markers. CONCLUSION: Practical measures of signal recovery are robust to linkage disequilibrium between a true causal variant and markers residing in the same genomic region. Given a limited sample size, it is possible to discover a phase transition by increasing the penalization; in this case a subset of the support may be recovered. Applying this approach to the GWAS analysis of height, we show that 70-100% of the selected markers are strongly correlated with height-associated markers identified by the GIANT Consortium.

Heritability and genetic correlations explained by common SNPs for metabolic syndrome traits.

We used a bivariate (multivariate) linear mixed-effects model to estimate the narrow-sense heritability (h(2)) and heritability explained by the common SNPs (h(g)(2)) for several metabolic syndrome (MetS) traits and the genetic correlation between pairs of traits for the Atherosclerosis Risk in Communities (ARIC) genome-wide association study (GWAS) population. MetS traits included body-mass index (BMI), waist-to-hip ratio (WHR), systolic blood pressure (SBP), fasting glucose (GLU), fasting insulin (INS), fasting trigylcerides (TG), and fasting high-density lipoprotein (HDL). We found the percentage of h(2) accounted for by common SNPs to be 58% of h(2) for height, 41% for BMI, 46% for WHR, 30% for GLU, 39% for INS, 34% for TG, 25% for HDL, and 80% for SBP. We confirmed prior reports for height and BMI using the ARIC population and independently in the Framingham Heart Study (FHS) population. We demonstrated that the multivariate model supported large genetic correlations between BMI and WHR and between TG and HDL. We also showed that the genetic correlations between the MetS traits are directly proportional to the phenotypic correlations.

A computational model for cerebral cortical dysfunction in autism spectrum disorders.

BACKGROUND: Perturbations to the microscopic level balance between synaptic excitation and inhibition and neuron organization in the cerebral cortex are suggested to underlie autism spectrum disorder (ASD) traits. The mechanism linking these perturbations to cognitive behaviors in ASD is unknown. This study strives to bridge this gap by generating clinically testable diagnostic and pharmacological predictions based on the effect of synaptic imbalance and neuron distribution on a computational local circuit model of the cerebral cortex. METHODS: We use a computational microscopic model of the cerebral cortex that incorporates N-methyl-D-aspartate and gamma-aminobutyric acid synaptic kinetics. We employ the model circuit during model tasks similar to visually guided and gap oculomotor saccade tasks and interpret qualitative model predictions of saccade hypometria and dysmetria. We consider the effects of varying the excitatory to inhibitory synaptic balance, neuron density, and neuron clustering in this model. RESULTS: An increase of synaptic excitation over synaptic inhibition results in increased hypometria and dysmetria. Similar effects by either reduced inhibition or increased excitation suggest that a variety of pharmacological compounds can be used for both screening and medical management. On the other hand, any change to the microscopic neuron anatomy that increases the effective maximum distance between excitatory neurons decreases hypometria but has no affect on dysmetria. CONCLUSIONS: Perturbations to a computational model of a local cerebral cortical circuit can account for saccade hypometria and dysmetria reported in ASD studies. This approach may provide a direct link between cerebral cortical function and ASD behaviors.

Prospective analysis of TEL/AML1-positive patients treated on Dana-Farber Cancer Institute Consortium Protocol 95-01.

In a retrospective analysis, we previously reported that children whose leukemia cells harbored the TEL/AML1 gene rearrangement have excellent outcomes. From 1996 to 2000, we conducted a prospective study to determine the incidence and outcomes of children with TEL/AML1-positive acute lymphoblastic leukemia (ALL). Children with newly diagnosed ALL were treated on DFCI ALL Consortium Protocol 95-01. Patients were risk stratified primarily by current National Cancer Institute (NCI)-Rome risk criteria. With a median follow-up of 5.2 years, the 5-year event-free survival for TEL/AML1-positive patients was 89% compared with 80% for TEL/AML1-negative B-precursor patients (P = .05). The 5-year overall survival rate was 97% among TEL/AML-positive patients compared with 89% among TEL/AML1-negative patients (P = .03). However, in a multivariable analysis, risk group (age and leukocyte count at diagnosis) and asparaginase treatment group, but not TEL/AML1 status, were found to be independent predictors of outcome. We conclude that TEL/AML1-positive patients have excellent outcomes, confirming our previous findings. However, factors such as age at diagnosis and presenting leukocyte count should be taken into consideration when treating this group of patients.

Acquired PTPN11 mutations occur rarely in adult patients with myelodysplastic syndromes and chronic myelomonocytic leukemia.

Myelodysplastic syndromes (MDS) are comprised of a heterogeneous group of stem cell disorders characterized by ineffective hematopoiesis and susceptibility to transform to acute myeloid leukemia. The molecular pathways underlying disease initiation and evolution are still largely unknown. We recently demonstrated that acquired mutations in PTPN11 are a major event in JMML and occur with variable prevalence in children with other hematologic malignancies, including MDS. Here, we investigated contribution of PTPN11 mutations to adult MDS and CMML pathogenesis. Our results indicate that PTPN11 lesions might play a role in adult MDS/CMML pathogenesis but do not represent a major molecular event.

Mutations in PTPN11 implicate the SHP-2 phosphatase in leukemogenesis.

The PTPN11 gene encodes SHP-2 (Src homology 2 domain-containing protein tyrosine Phosphatase), a nonreceptor tyrosine protein tyrosine phosphatase (PTPase) that relays signals from activated growth factor receptors to p21Ras (Ras) and other signaling molecules. Mutations in PTPN11 cause Noonan syndrome (NS), a developmental disorder characterized by cardiac and skeletal defects. NS is also associated with a spectrum of hematologic disorders, including juvenile myelomonocytic leukemia (JMML). To test the hypothesis that PTPN11 mutations might contribute to myeloid leukemogenesis, we screened the entire coding region for mutations in 51 JMML specimens and in selected exons from 60 patients with other myeloid malignancies. Missense mutations in PTPN11 were detected in 16 of 49 JMML specimens from patients without NS, but they were less common in other myeloid malignancies. RAS, NF1, and PTPN11 mutations are largely mutually exclusive in JMML, which suggests that mutant SHP-2 proteins deregulate myeloid growth through Ras. However, although Ba/F3 cells engineered to express leukemia-associated SHP-2 proteins cells showed enhanced growth factor-independent survival, biochemical analysis failed to demonstrate hyperactivation of the Ras effectors extracellular-regulated kinase (ERK) or Akt. We conclude that SHP-2 is an important cellular PTPase that is mutated in myeloid malignancies. Further investigation is required to clarify how these mutant proteins interact with Ras and other effectors to deregulate myeloid growth.