Overexpression of somatostatin receptors in patients with neuroendocrine neoplasms (NEN) is utilized for both diagnosis and treatment. Receptor density may reflect tumor differentiation and thus be associated with prognosis. Non-invasive visualization and quantification of somatostatin receptor density is possible by somatostatin receptor imaging (SRI) using positron emission tomography (PET). Recently, we introduced ⁶⁴Cu-DOTATATE for SRI and we hypothesized that uptake of this tracer could be associated with overall (OS) and progression-free survival (PFS). Methods: We evaluated patients with NEN that had a ⁶⁴Cu-DOTATATE PET/CT SRI performed in two prospective studies. Tracer uptake was determined as the maximal standardized uptake value (SUV_max) for each patient. Kaplan-Meier analysis with log-rank was used to determine the predictive value of ⁶⁴Cu-DOTATATE SUV_max for OS and PFS. Specificity, sensitivity and accuracy was calculated for prediction of outcome at 24 months after ⁶⁴Cu-DOTATATE PET/CT. Results: A total of 128 patients with NEN were included and followed for a median of 73 (1-112) months. During follow-up, 112 experienced disease progression and 69 patients died. The optimal cutoff for ⁶⁴Cu-DOTATATE SUV_max was 43.3 for prediction of PFS with a hazard ratio of 0.56 (95% CI: 0.38-0.84) for patients with SUV_max > 43.3. However, no significant cutoff was found for prediction of OS. In multiple Cox regression adjusted for age, sex, primary tumor site and tumor grade, the SUV_max cutoff hazard ratio was 0.50 (0.32-0.77) for PFS. The accuracy was moderate for predicting PFS (57%) at 24 months after ⁶⁴Cu-DOTATATE PET/CT. Conclusion: In this first study to report the association of ⁶⁴Cu-DOTATATE PET/CT and outcome in patients with NEN, tumor somatostatin receptor density visualized with ⁶⁴Cu-DOTATATE PET/CT was prognostic for PFS but not OS. However, the accuracy of prediction of PFS at 24 months after ⁶⁴Cu-DOTATATE PET/CT SRI was moderate limiting the value on an individual patient basis.

Bone is the most common site of distant metastatic spread in prostate adenocarcinoma. Prostate-specific membrane antigen uptake has been described in both benign and malignant bone lesions, which can lead to false-positive findings on ⁶⁸Ga-prostate-specific membrane antigen-11 positron emission tomography (⁶⁸Ga-PSMA-11 PET). The purpose of this study was to evaluate the diagnostic accuracy of ⁶⁸Ga-PSMA-11 PET for osseous prostate cancer metastases and improve bone uptake interpretation using semi-quantitative metrics. METHODS: 56 prostate cancer patients (18 pre-prostatectomy, 38 biochemical recurrence) who underwent ⁶⁸Ga-PSMA-11 PET/MRI or PET/CT examinations with osseous PSMA-ligand uptake were included in the study. Medical records were reviewed retrospectively by board-certified nuclear radiologists to determine true or false positivity based on a composite endpoint. For each avid osseous lesion, biological volume, size, PSMA-RADS rating, maximum standardized uptake value (SUV_max), and ratio of lesion SUV_max to liver, blood pool, and background bone SUV_max were measured. Differences between benign and malignant lesions were evaluated for statistical significance, and cut-off values for these parameters were determined to maximize diagnostic accuracy. RESULTS: Among 56 participants, 13 patients (22.8%) had false-positive osseous ⁶⁸Ga-PSMA-11 findings and 43 patients (76.8%) had true-positive osseous ⁶⁸Ga-PSMA-11 findings. Twenty-two patients (39%) had 1 osseous lesion, 18 (32%) had 2-4 lesions, and 16 (29%) had 5 or more lesions. Cut-off values resulting in statistically significant (p<0.005) differences between benign and malignant lesions were: PSMA-RADS ≥4, SUV_max ≥4.1, SUV_max ratio of lesion to blood pool ≥2.11, to liver ≥0.55, and to bone ≥4.4. These measurements corresponded to lesion-based ⁶⁸Ga-PSMA-11 PET lesion detection rate for malignancy of 80%, 93%, 89%, 21%, 89%, and a specificity of 73%, 73%, 73%, 93%, 60%, respectively. CONCLUSION: PSMA-RADS rating, SUV_max, and SUV_max ratio of lesion to blood pool can help differentiate benign from malignant lesions on ⁶⁸Ga-PSMA-11 PET. SUV_max ratio to blood pool above 2.2 is a reasonable parameter to support image interpretation and presented superior lesion detection rate and specificity when compared to visual interpretation by PSMA RADS. These parameters hold clinical value by improving diagnostic accuracy for metastatic prostate cancer on ⁶⁸Ga-PSMA-11 PET/MRI and PET/CT.

The presentation of peptides on class I human leukocyte antigen (HLA-I) molecules plays a central role in immune recognition of infected or malignant cells. In cancer, non-self HLA-I ligands can arise from many different alterations, including non-synonymous mutations, gene fusion, cancer-specific alternative mRNA splicing or aberrant post-translational modifications. Identifying HLA-I ligands remains a challenging task that requires either heavy experimental work for in vivo identification or optimized bioinformatics tools for accurate predictions. To date, no HLA-I ligand predictor includes post-translational modifications. To fill this gap, we curated phosphorylated HLA-I ligands from several immunopeptidomics studies (including six newly measured samples) covering 72 HLA-I alleles and retrieved a total of 2,066 unique phosphorylated peptides. We then expanded our motif deconvolution tool to identify precise binding motifs of phosphorylated HLA-I ligands. Our results reveal a clear enrichment of phosphorylated peptides among HLA-C ligands and demonstrate a prevalent role of both HLA-I motifs and kinase motifs on the presentation of phosphorylated peptides. These data further enabled us to develop and validate the first predictor of interactions between HLA-I molecules and phosphorylated peptides.

Acute myeloid leukemia (AML) is a clonal disorder arising from hematopoietic myeloid progenitors. Aberrantly activated tyrosine kinases (TK) are involved in leukemogenesis and are associated with poor treatment outcome. Kinase inhibitor (KI) treatment has shown promise in improving patient outcome in AML. However, inhibitor selection for patients is suboptimal.

In a preclinical effort to address KI selection, we analyzed a panel of 16 AML cell lines using phosphotyrosine (pY) enrichment-based, label-free phosphoproteomics. The Integrative Inferred Kinase Activity (INKA) algorithm was used to identify hyperphosphorylated, active kinases as candidates for KI treatment, and efficacy of selected KIs was tested.

Heterogeneous signaling was observed with between 241 and 2764 phosphopeptides detected per cell line. Of 4853 identified phosphopeptides with 4229 phosphosites, 4459 phosphopeptides (4430 pY) were linked to 3605 class I sites (3525 pY). INKA analysis in single cell lines successfully pinpointed driver kinases (PDGFRA, JAK2, KIT and FLT3) corresponding with activating mutations present in these cell lines. Furthermore, potential receptor tyrosine kinase (RTK) drivers, undetected by standard molecular analyses, were identified in four cell lines (FGFR1 in KG-1 and KG-1a, PDGFRA in Kasumi-3, and FLT3 in MM6). These cell lines proved highly sensitive to specific KIs. Six AML cell lines without a clear RTK driver showed evidence of MAPK1/3 activation, indicative of the presence of activating upstream RAS mutations. Importantly, FLT3 phosphorylation was demonstrated in two clinical AML samples with a FLT3 internal tandem duplication (ITD) mutation.

Our data show the potential of pY-phosphoproteomics and INKA analysis to provide insight in AML TK signaling and identify hyperactive kinases as potential targets for treatment in AML cell lines. These results warrant future investigation of clinical samples to further our understanding of TK phosphorylation in relation to clinical response in the individual patient.

1 April 2020

Here's an early look at how the Rays' 25-man roster could shape up on Opening Day.

Kristian Löbner
Mar 1, 2006; 55:792-797
Pathophysiology

With Spring Training set to kick off Tuesday, it feels like an opportune time to put together a way-too-early look at who might be with the Giants when they begin their regular season against the Padres on March 28.

At some point over the next six weeks, injuries, improvement or regression will inevitably alter the plan nearly every Major League club brings into Spring Training. The Braves must decide exactly how to round out their rotation and bullpen, but for now it looks like they won't have any position-player battles. Here's the first prediction of how Atlanta's Opening Day roster might look.

Here's an early look at how the Orioles' 25-man roster could shape up on Opening Day.

The ability to switch fuels for oxidation in response to changes in macronutrient composition of diet (metabolic flexibility) may be informative of individuals’ susceptibility to weight gain. Seventy-nine healthy, weight-stable participants underwent 24-h assessments of energy expenditure and respiratory quotient (RQ) in a whole-room calorimeter during energy balance (EBL) (50% carbohydrate, 30% fat) and then during 24-h fasting and three 200% overfeeding diets in a crossover design. Metabolic flexibility was defined as the change in 24-h RQ from EBL during fasting and standard overfeeding (STOF) (50% carbohydrate, 30% fat), high-fat overfeeding (HFOF) (60% fat, 20% carbohydrate), and high-carbohydrate overfeeding (HCOF) (75% carbohydrate, 5% fat) diets. Free-living weight change was assessed after 6 and 12 months. Compared with EBL, RQ decreased on average by 9% during fasting and by 4% during HFOF but increased by 4% during STOF and by 8% during HCOF. A smaller decrease in RQ, reflecting a smaller increase in lipid oxidation rate, during HFOF but not during the other diets predicted greater weight gain at both 6 and 12 months. An impaired metabolic flexibility to acute HFOF can identify individuals prone to weight gain, indicating that an individual’s capacity to oxidize dietary fat is a metabolic determinant of weight change.

There won't be many roster battles when the Astros open camp later this week, and they won't be decided until the final days of the team's stay in West Palm Beach, Fla. Houston has an open competition for its fifth starter spot heading into Spring Training, and the club also has to sort out a crowded outfield and the final spots in the bullpen.

There comes a tipping point in all campaigns when the evidence is overwhelming and the only way to proceed is with action. According to David Williams, it’s time to tackle the disproportionate effects of race on patients in the UK. David Williams, from Harvard University, developed the Everyday Discrimination Scale that, in 1997, launched a new...

Dilys J. Freeman
May 1, 2002; 51:1596-1600
Complications

Andreas Festa
Apr 1, 2002; 51:1131-1137
Complications

This study aims to model genetic, immunologic, metabolomics, and proteomic biomarkers for development of islet autoimmunity (IA) and progression to type 1 diabetes in a prospective high-risk cohort. We studied 67 children: 42 who developed IA (20 of 42 progressed to diabetes) and 25 control subjects matched for sex and age. Biomarkers were assessed at four time points: earliest available sample, just prior to IA, just after IA, and just prior to diabetes onset. Predictors of IA and progression to diabetes were identified across disparate sources using an integrative machine learning algorithm and optimization-based feature selection. Our integrative approach was predictive of IA (area under the receiver operating characteristic curve [AUC] 0.91) and progression to diabetes (AUC 0.92) based on standard cross-validation (CV). Among the strongest predictors of IA were change in serum ascorbate, 3-methyl-oxobutyrate, and the PTPN22 (rs2476601) polymorphism. Serum glucose, ADP fibrinogen, and mannose were among the strongest predictors of progression to diabetes. This proof-of-principle analysis is the first study to integrate large, diverse biomarker data sets into a limited number of features, highlighting differences in pathways leading to IA from those predicting progression to diabetes. Integrated models, if validated in independent populations, could provide novel clues concerning the pathways leading to IA and type 1 diabetes.

WASHINGTON, CMC – The World Bank has predicted the sharpest decline of remittances to Latin America and the Caribbean, saying that global remittances on a whole are projected to fall by about 20 percent in 2020 due to the economic crisis...

Here's an early look at how the Reds' 25-man roster could shape up on Opening Day.

Here's an early look at how the D-backs' 25-man roster could shape up on Opening Day.

The present study examined the predictive values of amyloid PET, ¹⁸F-FDG PET, and nonimaging predictors (alone and in combination) for development of Alzheimer dementia (AD) in a large population of patients with mild cognitive impairment (MCI). Methods: The study included 319 patients with MCI from the Alzheimer Disease Neuroimaging Initiative database. In a derivation dataset (n = 159), the following Cox proportional-hazards models were constructed, each adjusted for age and sex: amyloid PET using ¹⁸F-florbetapir (pattern expression score of an amyloid-β AD conversion–related pattern, constructed by principle-components analysis); ¹⁸F-FDG PET (pattern expression score of a previously defined ¹⁸F-FDG–based AD conversion–related pattern, constructed by principle-components analysis); nonimaging (functional activities questionnaire, apolipoprotein E, and mini-mental state examination score); ¹⁸F-FDG PET + amyloid PET; amyloid PET + nonimaging; ¹⁸F-FDG PET + nonimaging; and amyloid PET + ¹⁸F-FDG PET + nonimaging. In a second step, the results of Cox regressions were applied to a validation dataset (n = 160) to stratify subjects according to the predicted conversion risk. Results: On the basis of the independent validation dataset, the ¹⁸F-FDG PET model yielded a significantly higher predictive value than the amyloid PET model. However, both were inferior to the nonimaging model and were significantly improved by the addition of nonimaging variables. The best prediction accuracy was reached by combining ¹⁸F-FDG PET, amyloid PET, and nonimaging variables. The combined model yielded 5-y free-of-conversion rates of 100%, 64%, and 24% for the low-, medium- and high-risk groups, respectively. Conclusion: ¹⁸F-FDG PET, amyloid PET, and nonimaging variables represent complementary predictors of conversion from MCI to AD. Especially in combination, they enable an accurate stratification of patients according to their conversion risks, which is of great interest for patient care and clinical trials.

Members Event Webinar

Event participants

Professor Matthew Goodwin, Visiting Senior Fellow, Europe Programme, Chatham House
Professor David Cutts, Associate Fellow, Europe Programme, Chatham House

OBJECTIVE

Type 2 diabetes mellitus (T2DM) is associated with dyslipidemia, but the detailed alterations in lipid species preceding the disease are largely unknown. We aimed to identify plasma lipids associated with development of T2DM and investigate their associations with lifestyle.

OBJECTIVE

To evaluate the respective contributions of short-term glycemic variability and mean daily glucose (MDG) concentration to the risk of hypoglycemia in type 1 diabetes.

OBJECTIVE

This study evaluated the ability of the Building, Relating, Assessing, and Validating Outcomes (BRAVO) risk engine to accurately project cardiovascular outcomes in three major clinical trials—BI 10773 (Empagliflozin) Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME), Canagliflozin Cardiovascular Assessment Study (CANVAS), and Dapagliflozin Effect on Cardiovascular Events–Thrombolysis in Myocardial Infarction (DECLARE-TIMI 58) trial—on sodium–glucose cotransporter 2 inhibitors (SGLT2is) to treat patients with type 2 diabetes.

OBJECTIVE

We analyzed data from inpatients with diabetes admitted to a large university hospital to predict the risk of hypoglycemia through the use of machine learning algorithms.

Based on the newest forecasting models, AccuWeather forecasters have extended the upper range of hurricanes predicted for the Atlantic hurricane season.

OBJECTIVE

To construct and internally validate prediction models to estimate the risk of long-term end-organ complications and mortality in patients with type 2 diabetes and obesity that can be used to inform treatment decisions for patients and practitioners who are considering metabolic surgery.

Neurosciences.

19th century.

Research from REL Midwest examines the student characteristics associated with early college success in Indiana, with a focus on financial aid.

François Bachoc, Baptiste Broto, Fabrice Gamboa, Jean-Michel Loubes.

Source: Electronic Journal of Statistics, Volume 14, Number 1, 503--546.

Abstract:
In the framework of the supervised learning of a real function defined on an abstract space $mathcal{X}$, Gaussian processes are widely used. The Euclidean case for $mathcal{X}$ is well known and has been widely studied. In this paper, we explore the less classical case where $mathcal{X}$ is the non commutative finite group of permutations (namely the so-called symmetric group $S_{N}$). We provide an application to Gaussian process based optimization of Latin Hypercube Designs. We also extend our results to the case of partial rankings.

Assaf Rabinowicz, Saharon Rosset.

Source: Electronic Journal of Statistics, Volume 14, Number 1, 272--301.

Abstract:
Common model selection criteria, such as $AIC$ and its variants, are based on in-sample prediction error estimators. However, in many applications involving predicting at interpolation and extrapolation points, in-sample error does not represent the relevant prediction error. In this paper new prediction error estimators, $tAI$ and $Loss(w_{t})$ are introduced. These estimators generalize previous error estimators, however are also applicable for assessing prediction error in cases involving interpolation and extrapolation. Based on these prediction error estimators, two model selection criteria with the same spirit as $AIC$ and Mallow’s $C_{p}$ are suggested. The advantages of our suggested methods are demonstrated in a simulation and a real data analysis of studies involving interpolation and extrapolation in linear mixed model and Gaussian process regression.

Ben Jones, Andreas Artemiou, Bing Li.

Source: Electronic Journal of Statistics, Volume 14, Number 1, 1--23.

Abstract:
We give a probabilistic analysis of a phenomenon in statistics which, until recently, has not received a convincing explanation. This phenomenon is that the leading principal components tend to possess more predictive power for a response variable than lower-ranking ones despite the procedure being unsupervised. Our result, in its most general form, shows that the phenomenon goes far beyond the context of linear regression and classical principal components — if an arbitrary distribution for the predictor $X$ and an arbitrary conditional distribution for $Yvert X$ are chosen then any measureable function $g(Y)$, subject to a mild condition, tends to be more correlated with the higher-ranking kernel principal components than with the lower-ranking ones. The “arbitrariness” is formulated in terms of unitary invariance then the tendency is explicitly quantified by exploring how unitary invariance relates to the Cauchy distribution. The most general results, for technical reasons, are shown for the case where the kernel space is finite dimensional. The occurency of this tendency in real world databases is also investigated to show that our results are consistent with observation.

Tomáš Rubín, Victor M. Panaretos.

Source: Electronic Journal of Statistics, Volume 14, Number 1, 1137--1210.

Abstract:
Functional time series analysis, whether based on time or frequency domain methodology, has traditionally been carried out under the assumption of complete observation of the constituent series of curves, assumed stationary. Nevertheless, as is often the case with independent functional data, it may well happen that the data available to the analyst are not the actual sequence of curves, but relatively few and noisy measurements per curve, potentially at different locations in each curve’s domain. Under this sparse sampling regime, neither the established estimators of the time series’ dynamics nor their corresponding theoretical analysis will apply. The subject of this paper is to tackle the problem of estimating the dynamics and of recovering the latent process of smooth curves in the sparse regime. Assuming smoothness of the latent curves, we construct a consistent nonparametric estimator of the series’ spectral density operator and use it to develop a frequency-domain recovery approach, that predicts the latent curve at a given time by borrowing strength from the (estimated) dynamic correlations in the series across time. This new methodology is seen to comprehensively outperform a naive recovery approach that would ignore temporal dependence and use only methodology employed in the i.i.d. setting and hinging on the lag zero covariance. Further to predicting the latent curves from their noisy point samples, the method fills in gaps in the sequence (curves nowhere sampled), denoises the data, and serves as a basis for forecasting. Means of providing corresponding confidence bands are also investigated. A simulation study interestingly suggests that sparse observation for a longer time period may provide better performance than dense observation for a shorter period, in the presence of smoothness. The methodology is further illustrated by application to an environmental data set on fair-weather atmospheric electricity, which naturally leads to a sparse functional time series.

John J. Dziak, Donna L. Coffman, Matthew Reimherr, Justin Petrovich, Runze Li, Saul Shiffman, Mariya P. Shiyko.

Source: Statistics Surveys, Volume 13, 150--180.

Abstract:
Researchers are sometimes interested in predicting a distal or external outcome (such as smoking cessation at follow-up) from the trajectory of an intensively recorded longitudinal variable (such as urge to smoke). This can be done in a semiparametric way via scalar-on-function regression. However, the resulting fitted coefficient regression function requires special care for correct interpretation, as it represents the joint relationship of time points to the outcome, rather than a marginal or cross-sectional relationship. We provide practical guidelines, based on experience with scientific applications, for helping practitioners interpret their results and illustrate these ideas using data from a smoking cessation study.

Jonathan R. Bradley, Noel Cressie, Tao Shi.

Source: Statistics Surveys, Volume 10, 100--131.

Abstract:
In this article, we review and compare a number of methods of spatial prediction, where each method is viewed as an algorithm that processes spatial data. To demonstrate the breadth of available choices, we consider both traditional and more-recently-introduced spatial predictors. Specifically, in our exposition we review: traditional stationary kriging, smoothing splines, negative-exponential distance-weighting, fixed rank kriging, modified predictive processes, a stochastic partial differential equation approach, and lattice kriging. This comparison is meant to provide a service to practitioners wishing to decide between spatial predictors. Hence, we provide technical material for the unfamiliar, which includes the definition and motivation for each (deterministic and stochastic) spatial predictor. We use a benchmark dataset of $mathrm{CO}_{2}$ data from NASA’s AIRS instrument to address computational efficiencies that include CPU time and memory usage. Furthermore, the predictive performance of each spatial predictor is assessed empirically using a hold-out subset of the AIRS data.

Aki Vehtari, Janne Ojanen.

Source: Statistics Surveys, Volume 8, , 1--1.

Abstract:
Errata for “A survey of Bayesian predictive methods for model assessment, selection and comparison” by A. Vehtari and J. Ojanen, Statistics Surveys , 6 (2012), 142–228. doi:10.1214/12-SS102.

Aki Vehtari, Janne Ojanen

Source: Statist. Surv., Volume 6, 142--228.

Abstract:
To date, several methods exist in the statistical literature for model assessment, which purport themselves specifically as Bayesian predictive methods. The decision theoretic assumptions on which these methods are based are not always clearly stated in the original articles, however. The aim of this survey is to provide a unified review of Bayesian predictive model assessment and selection methods, and of methods closely related to them. We review the various assumptions that are made in this context and discuss the connections between different approaches, with an emphasis on how each method approximates the expected utility of using a Bayesian model for the purpose of predicting future data.

Bertrand Clarke, Jennifer Clarke

Source: Statist. Surv., Volume 6, 1--73.

Abstract:
We review predictive techniques from several traditional branches of statistics. Starting with prediction based on the normal model and on the empirical distribution function, we proceed to techniques for various forms of regression and classification. Then, we turn to time series, longitudinal data, and survival analysis. Our focus throughout is on the mechanics of prediction more than on the properties of predictors.

Early detection of patients vulnerable to infections acquired in the hospital environment is a challenge in current health systems given the impact that such infections have on patient mortality and healthcare costs. This work is focused on both the identification of risk factors and the prediction of healthcare-associated infections in intensive-care units by means of machine-learning methods. The aim is to support decision making addressed at reducing the incidence rate of infections. In this field, it is necessary to deal with the problem of building reliable classifiers from imbalanced datasets. We propose a clustering-based undersampling strategy to be used in combination with ensemble classifiers. A comparative study with data from 4616 patients was conducted in order to validate our proposal. We applied several single and ensemble classifiers both to the original dataset and to data preprocessed by means of different resampling methods. The results were analyzed by means of classic and recent metrics specifically designed for imbalanced data classification. They revealed that the proposal is more efficient in comparison with other approaches.

In the variational relevance vector machine, the gamma distribution is representative as a hyperprior over the noise precision of automatic relevance determination prior. Instead of the gamma hyperprior, we propose to use the inverse gamma hyperprior with a shape parameter close to zero and a scale parameter not necessary close to zero. This hyperprior is associated with the concept of a weakly informative prior. The effect of this hyperprior is investigated through regression to non-homogeneous data. Because it is difficult to capture the structure of such data with a single kernel function, we apply the multiple kernel method, in which multiple kernel functions with different widths are arranged for input data. We confirm that the degrees of freedom in a model is controlled by adjusting the scale parameter and keeping the shape parameter close to zero. A candidate for selecting the scale parameter is the predictive information criterion. However the estimated model using this criterion seems to cause over-fitting. This is because the multiple kernel method makes the model a situation where the dimension of the model is larger than the data size. To select an appropriate scale parameter even in such a situation, we also propose an extended prediction information criterion. It is confirmed that a multiple kernel relevance vector regression model with good predictive accuracy can be obtained by selecting the scale parameter minimizing extended prediction information criterion.

Effective property prediction methods can help accelerate the search for COVID-19 antivirals either through accurate in-silico screens or by effectively guiding on-going at-scale experimental efforts. However, existing prediction tools have limited ability to accommodate scarce or fragmented training data currently available. In this paper, we introduce a novel approach to learn predictors that can generalize or extrapolate beyond the heterogeneous data. Our method builds on and extends recently proposed invariant risk minimization, adaptively forcing the predictor to avoid nuisance variation. We achieve this by continually exercising and manipulating latent representations of molecules to highlight undesirable variation to the predictor. To test the method we use a combination of three data sources: SARS-CoV-2 antiviral screening data, molecular fragments that bind to SARS-CoV-2 main protease and large screening data for SARS-CoV-1. Our predictor outperforms state-of-the-art transfer learning methods by significant margin. We also report the top 20 predictions of our model on Broad drug repurposing hub.

Edgar Dobriban, William Leeb, Amit Singer.

Source: The Annals of Statistics, Volume 48, Number 1, 491--513.

Abstract:
We consider the linearly transformed spiked model , where the observations $Y_{i}$ are noisy linear transforms of unobserved signals of interest $X_{i}$: egin{equation*}Y_{i}=A_{i}X_{i}+varepsilon_{i},end{equation*} for $i=1,ldots ,n$. The transform matrices $A_{i}$ are also observed. We model the unobserved signals (or regression coefficients) $X_{i}$ as vectors lying on an unknown low-dimensional space. Given only $Y_{i}$ and $A_{i}$ how should we predict or recover their values? The naive approach of performing regression for each observation separately is inaccurate due to the large noise level. Instead, we develop optimal methods for predicting $X_{i}$ by “borrowing strength” across the different samples. Our linear empirical Bayes methods scale to large datasets and rely on weak moment assumptions. We show that this model has wide-ranging applications in signal processing, deconvolution, cryo-electron microscopy, and missing data with noise. For missing data, we show in simulations that our methods are more robust to noise and to unequal sampling than well-known matrix completion methods.

Mohamad Elmasri, Maxwell J. Farrell, T. Jonathan Davies, David A. Stephens.

Source: The Annals of Applied Statistics, Volume 14, Number 1, 221--240.

Abstract:
Identifying undocumented or potential future interactions among species is a challenge facing modern ecologists. Recent link prediction methods rely on trait data; however, large species interaction databases are typically sparse and covariates are limited to only a fraction of species. On the other hand, evolutionary relationships, encoded as phylogenetic trees, can act as proxies for underlying traits and historical patterns of parasite sharing among hosts. We show that, using a network-based conditional model, phylogenetic information provides strong predictive power in a recently published global database of host-parasite interactions. By scaling the phylogeny using an evolutionary model, our method allows for biological interpretation often missing from latent variable models. To further improve on the phylogeny-only model, we combine a hierarchical Bayesian latent score framework for bipartite graphs that accounts for the number of interactions per species with host dependence informed by phylogeny. Combining the two information sources yields significant improvement in predictive accuracy over each of the submodels alone. As many interaction networks are constructed from presence-only data, we extend the model by integrating a correction mechanism for missing interactions which proves valuable in reducing uncertainty in unobserved interactions.

Elizabeth Lorenzi, Ricardo Henao, Katherine Heller.

Source: The Annals of Applied Statistics, Volume 13, Number 4, 2637--2661.

Abstract:
Nearly a third of all surgeries performed in the United States occur for patients over the age of 65; these older adults experience a higher rate of postoperative morbidity and mortality. To improve the care for these patients, we aim to identify and characterize high risk geriatric patients to send to a specialized perioperative clinic while leveraging the overall surgical population to improve learning. To this end, we develop a hierarchical infinite latent factor model (HIFM) to appropriately account for the covariance structure across subpopulations in data. We propose a novel Hierarchical Dirichlet Process shrinkage prior on the loadings matrix that flexibly captures the underlying structure of our data while sharing information across subpopulations to improve inference and prediction. The stick-breaking construction of the prior assumes an infinite number of factors and allows for each subpopulation to utilize different subsets of the factor space and select the number of factors needed to best explain the variation. We develop the model into a latent factor regression method that excels at prediction and inference of regression coefficients. Simulations validate this strong performance compared to baseline methods. We apply this work to the problem of predicting surgical complications using electronic health record data for geriatric patients and all surgical patients at Duke University Health System (DUHS). The motivating application demonstrates the improved predictive performance when using HIFM in both area under the ROC curve and area under the PR Curve while providing interpretable coefficients that may lead to actionable interventions.

Silvia Metelli, Nicholas Heard.

Source: The Annals of Applied Statistics, Volume 13, Number 4, 2586--2610.

Abstract:
Monitoring computer network traffic for anomalous behaviour presents an important security challenge. Arrivals of new edges in a network graph represent connections between a client and server pair not previously observed, and in rare cases these might suggest the presence of intruders or malicious implants. We propose a Bayesian model and anomaly detection method for simultaneously characterising existing network structure and modelling likely new edge formation. The method is demonstrated on real computer network authentication data and successfully identifies some machines which are known to be compromised.