Search

It is crucial to assess the predictive performance of a model in order to establish its practicality and relevance in real-world scenarios, particularly for high-dimensional data analysis. Among data splitting or resampling methods, cross-validation (CV) is extensively used for several tasks such as estimating the prediction error, tuning the regularization parameter, and selecting the most suitable predictive model among competing alternatives. The K-fold cross-validation is a popular CV method but its limitation is that the risk estimates are highly dependent on the partitioning of the data (for training and testing). Here, the issues regarding the reproducibility of the K-fold CV estimator is demonstrated in hypothesis testing wherein different partitions lead to notably disparate conclusions. This study presents an alternative novel predictive performance test and valid confidence intervals based on exhaustive nested cross-validation for determining the difference in prediction error between two model-fitting algorithms. A naive implementation of the exhaustive nested cross-validation is computationally costly. Here, we address concerns regarding computational complexity by devising a computationally tractable closed-form expression for the proposed cross-validation estimator using ridge regularization. Our study also investigates strategies aimed at enhancing statistical power within high-dimensional scenarios while controlling the Type I error rate. To illustrate the practical utility of our method, we apply it to an RNA sequencing study and demonstrate its effectiveness in the context of biological data analysis., Comment: 49 pages, 7 figures

Molecular and genomic technological advancements have greatly enhanced our understanding of biological processes by allowing us to quantify key biological variables such as gene expression, protein levels, and microbiome compositions. These breakthroughs have enabled us to achieve increasingly higher levels of resolution in our measurements, exemplified by our ability to comprehensively profile biological information at the single-cell level. However, the analysis of such data faces several critical challenges: limited number of individuals, non-normality, potential dropouts, outliers, and repeated measurements from the same individual. In this article, we propose a novel method, which we call U-statistic based latent variable (ULV). Our proposed method takes advantage of the robustness of rank-based statistics and exploits the statistical efficiency of parametric methods for small sample sizes. It is a computationally feasible framework that addresses all the issues mentioned above simultaneously. An additional advantage of ULV is its flexibility in modeling various types of single-cell data, including both RNA and protein abundance. The usefulness of our method is demonstrated in two studies: a single-cell proteomics study of acute myelogenous leukemia (AML) and a single-cell RNA study of COVID-19 symptoms. In the AML study, ULV successfully identified differentially expressed proteins that would have been missed by the pseudobulk version of the Wilcoxon rank-sum test. In the COVID-19 study, ULV identified genes associated with covariates such as age and gender, and genes that would be missed without adjusting for covariates. The differentially expressed genes identified by our method are less biased toward genes with high expression levels. Furthermore, ULV identified additional gene pathways likely contributing to the mechanisms of COVID-19 severity.

We propose a statistical framework to investigate whether a given subpopulation lies between two other subpopulations in a multivariate feature space. This methodology is motivated by a biological question from a collaborator: Is a newly discovered cell type between two known types in several given features? We propose two in-betweenness indices (IBI) to quantify the in-betweenness exhibited by a random triangle formed by the summary statistics of the three subpopulations. Statistical inference methods are provided for triangle shape and IBI metrics. The application of our methods is demonstrated in three examples: the classic Iris data set, a study of risk of relapse across three breast cancer subtypes, and the motivating neuronal cell data with measured electrophysiological features.

With the increasing popularity of calcium imaging data in neuroscience research, methods for analyzing calcium trace data are critical to address various questions. The observed calcium traces are either analyzed directly or deconvolved to spike trains to infer neuronal activities. When both approaches are applicable, it is unclear whether deconvolving calcium traces is a necessary step. In this article, we compare the performance of using calcium traces or their deconvolved spike trains for three common analyses: clustering, principal component analysis (PCA), and population decoding. Our simulations and applications to real data suggest that the estimated spike data outperform calcium trace data for both clustering and PCA. Although calcium trace data show higher predictability than spike data at each time point, spike history or cumulative spike counts is comparable to or better than calcium traces in population decoding.

We propose a ridge-penalized adaptive Mantel test (AdaMant) for evaluating the association of two high-dimensional sets of features. By introducing a ridge penalty, AdaMant tests the association across many metrics simultaneously. We demonstrate how ridge penalization bridges Euclidean and Mahalanobis distances and their corresponding linear models from the perspective of association measurement and testing. This result is not only theoretically interesting but also has important implications in penalized hypothesis testing, especially in high dimensional settings such as imaging genetics. Applying the proposed method to an imaging genetic study of visual working memory in health adults, we identified interesting associations of brain connectivity (measured by EEG coherence) with selected genetic features.

Optical imaging of genetically encoded calcium indicators is a powerful tool to record the activity of a large number of neurons simultaneously over a long period of time from freely behaving animals. However, determining the exact time at which a neuron spikes and estimating the underlying firing rate from calcium fluorescence data remains challenging, especially for calcium imaging data obtained from a longitudinal study. We propose a multi-trial time-varying $\ell_0$ penalized method to jointly detect spikes and estimate firing rates by robustly integrating evolving neural dynamics across trials. Our simulation study shows that the proposed method performs well in both spike detection and firing rate estimation. We demonstrate the usefulness of our method on calcium fluorescence trace data from two studies, with the first study showing differential firing rate functions between two behaviors and the second study showing evolving firing rate function across trials due to learning.

Recent advancements in miniaturized fluorescence microscopy have made it possible to investigate neuronal responses to external stimuli in awake behaving animals through the analysis of intra-cellular calcium signals. An on-going challenge is deconvolving the temporal signals to extract the spike trains from the noisy calcium signals' time-series. In this manuscript, we propose a nested Bayesian finite mixture specification that allows the estimation of spiking activity and, simultaneously, reconstructing the distributions of the calcium transient spikes' amplitudes under different experimental conditions. The proposed model leverages two nested layers of random discrete mixture priors to borrow information between experiments and discover similarities in the distributional patterns of neuronal responses to different stimuli. Furthermore, the spikes' intensity values are also clustered within and between experimental conditions to determine the existence of common (recurring) response amplitudes. Simulation studies and the analysis of a data set from the Allen Brain Observatory show the effectiveness of the method in clustering and detecting neuronal activities., Comment: 18 pages, 5 figures

We propose a two-stage approach Spec PC-CP to identify change points in multivariate time series. In the first stage, we obtain a low-dimensional summary of the high-dimensional time series by Spectral Principal Component Analysis (Spec-PCA). In the second stage, we apply cumulative sum-type test on the Spectral PCA component using a binary segmentation algorithm. Compared with existing approaches, the proposed method is able to capture the lead-lag relationship in time series. Our simulations demonstrate that the Spec PC-CP method performs significantly better than competing methods for detecting change points in high-dimensional time series. The results on epileptic seizure EEG data and stock data also indicate that our new method can efficiently {detect} change points corresponding to the onset of the underlying events.

Mantel's test (MT) for association is conducted by testing the linear relationship of similarity of all pairs of subjects between two observational domains. Motivated by applications to neuroimaging and genetics data, and following the succes of shrinkage and kernel methods for prediction with high-dimensional data, we here introduce the adaptive Mantel test as an extension of the MT. By utilizing kernels and penalized similarity measures, the adaptive Mantel test is able to achieve higher statistical power relative to the classical MT in many settings. Furthermore, the adaptive Mantel test is designed to simultaneously test over multiple similarity measures such that the correct type I error rate under the null hypothesis is maintained without the need to directly adjust the significance threshold for multiple testing. The performance of the adaptive Mantel test is evaluated on simulated data, and is used to investigate associations between genetics markers related to Alzheimer's Disease and heatlhy brain physiology with data from a working memory study of 350 college students from Beijing Normal University.

Brain signal data are inherently big: massive in amount, complex in structure, and high in dimensions. These characteristics impose great challenges for statistical inference and learning. Here we review several key challenges, discuss possible solutions, and highlight future research directions.