Hao Feng

Single-cell DNA methylation profiling technology captures novel epigenetic data modality but are challenging to analyze because of their heterogeneity, high dimensionality, and ultra-sparsity. Here we present SMORE ( S ingle-cell M ethyl O me R eduction and E mbedding), a computational method for joint dimensionality reduction and cell population discovery dedicated to single cell DNA methylation data. SMORE operates on a Bayesian framework that converts methylation proportions into ordered methylation states and jointly infers a low-dimensional representation, cell populations and their number. By using low-rank latent Gaussian factorization and adopting a mixture-of-finite-mixtures prior on latent cell scores, SMORE infers cell assignments without requiring a prespecified cluster number, and propagates uncertainty from methylation measurements to cell assignments. Across simulations spanning varying sample sizes, population imbalance, signal strengths and model misspecification, SMORE accurately recovered latent population structure and outperformed existing methods. Applied to human single-cell methylation datasets from lung, peripheral blood and primary motor cortex, SMORE recovered biologically supported cell population structures. SMORE provides an uncertainty-aware framework for dimension reduction and population discovery for single-cell methylomes.

Recent advancements in single-cell DNA methylation (scDNAm) sequencing technologies have enabled the profiling of epigenetic landscapes at unprecedented resolution, offering insights into cellular heterogeneity, differentiation and evolution. Trajectory inference, which orders cells along pseudotime, allows researchers to track genomics changes across continuous cell states and identify key loci exhibiting differential methylation. However, no methods currently exist to model methylation changes along pseudotime in scDNAm data. Here, we present a hierarchical Bayesian framework for scDNAm data analysis. Our method, named mist (methylation inference for single-cell along trajectory), models stage-specific biological variations, identifies genomic features with significant methylation changes along pseudotime, and performs Differential Methylation (DM) analysis across phenotypical groups. Simulations demonstrate its superior accuracy in detecting DM genes along pseudotime compared to existing methods. Applied to multi-omics datasets of mouse embryonic development and developing human brain, mist identifies key developmental regulators, whose methylation patterns align with lineage transitions. mist is publicly available as an R/Bioconductor package.

Alzheimer’s disease pathology (ADP) and Lewy body pathology (LBP) are traditionally associated with distinct cognitive profiles. However, growing evidence highlights the role of behavioral and psychological symptoms of dementia (BPSD) in shaping clinical presentations. The combined influence of cognitive and behavioral symptoms across neuropathologically confirmed ADP, LBP, and mixed AD-LBP has not been systematically examined. This study aimed to identify clinically meaningful subtypes by jointly analyzing cognitive performance and BPSD profiles in individuals with autopsy-confirmed dementia pathology. This retrospective longitudinal cohort study used data from the National Alzheimer Coordinating Center (NACC), collected across multiple U.S. Alzheimer’s Disease Research Centers. Participants had a Clinical Dementia Rating (CDR) Global score ≤1 at baseline and autopsy-confirmed ADP, LBP, or mixed AD-LBP. Cognitive outcomes included standardized tests of memory, executive function, and language. BPSD were assessed using the Neuropsychiatric Inventory Questionnaire (NPI-Q), which captures ten symptom domains: agitation, apathy, depression, delusions, disinhibition, auditory and visual hallucinations, irritability, personality change, and REM sleep behavior disorder. Cluster analysis was applied to identify subtypes based on combined cognitive and BPSD data. The study included 1,028 participants (mean age at baseline 76.4 years; 47.6% female): 521 with ADP, 96 with LBP, and 411 with mixed AD-LBP. A three-cluster clinical subtype (CS) solution best fit the data. The most symptomatic group (CS-3) had the youngest age at first visit (mean 72.1 years), the highest BPSD burden, and the fastest cognitive and functional decline across ADP and AD-LBP groups. CS-1 and CS-2 exhibited milder early cognitive impairment and lower BPSD burden. Within ADP and AD-LBP, CS-2 showed slower progression than CS-1, fewer APOE ε4 carriers (41% vs. 58%), and better memory scores, despite reporting a higher frequency of agitation. These findings reveal distinct clinical subtypes that cut across traditional pathological boundaries, emphasizing the need to incorporate both cognitive and behavioral features into early dementia characterization. This multidimensional approach can improve guide personalized prognosis and care planning and enhance clinical trial design by considering disease heterogeneity. The study supports integrated clinical profiling as important factor in robust evaluation of dementia outcomes.

In the era of big data, increasing availability of data makes combining different data sources to obtain more accurate estimations a popular topic. However, the development of data integration is often hindered by the heterogeneity in data forms across studies. In this paper, we focus on a case in survival analysis where we have primary study data with a continuous time-to-event outcome and complete covariate measurements, while the data from an external study contain an outcome observed at regular intervals, and only a subset of covariates is measured. To incorporate external information while accounting for the different data forms, we posit working models and obtain informative weights by empirical likelihood, which will be used to construct a weighted estimator in the main analysis. We have established the theory demonstrating that the new estimator has higher estimation efficiency compared to the conventional ones, and this advantage is robust to working model misspecification, as confirmed in our simulation studies. To assess its utility, we apply our method to accommodate data from the National Alzheimer’s Coordinating Center to improve the analysis of the Alzheimer’s Disease Neuroimaging Initiative Phase 1 study.

Detection and Dissection of Anomalous Tissue Domains (DDATD) from multi-sample spatial transcriptomics (ST) data provides unprecedented opportunities to characterize anomalous tissue domains (ATDs), revealing both population-level and individual-specific pathogenic factors for understanding pathogenic heterogeneities behind diseases. However, no current methods can perform de novo DDATD from ST data, especially in the multi-sample context. Here, we introduce STANDS, an innovative framework based on Generative Adversarial Networks which integrates three core tasks in multi-sample DDATD: detecting, aligning, and subtyping ATDs. STANDS incorporates multimodal-learning, transfer-learning, and style-transfer techniques to effectively address major challenges in multi-sample DDATD, including complications caused by unalignable ATDs, under-utilization of multimodal information, and scarcity of normal ST datasets necessary for comparative analysis. Extensive benchmarks from diverse datasets demonstrate STAND’s superiority in identifying both common and individual-specific ATDs and further dissecting them into biologically distinct subdomains. STANDS also provides clues to developing ATDs visually indistinguishable from surrounding normal tissues.

Recent methodology advances in computational signal deconvolution have enabled bulk transcriptome data analysis at a finer cell-type level. Through deconvolution, identifying cell-type-specific differentially expressed (csDE) genes is drawing increasing attention in clinical applications. However, researchers still face a number of difficulties in adopting csDE genes detection methods in practice, especially in their experimental design. Here we present cypress, the first experimental design and statistical power analysis tool in csDE genes identification. This tool can reliably model purified cell-type-specific (CTS) profiles, cell-type compositions, biological and technical variations, offering a high-fidelity simulator for bulk RNA-seq convolution and deconvolution. cypress conducts simulation and evaluates the impact of multiple influencing factors, by various statistical metrics, to help researchers optimize experimental design and conduct power analysis. cypress is an open-source R/Bioconductor package at https://bioconductor.org/packages/cypress/.

In clinical and observational studies, secondary outcomes are frequently collected alongside the primary outcome for each subject, yet their potential to improve the analysis efficiency remains underutilized. Moreover, missing data, commonly encountered in practice, can introduce bias to estimates if not appropriately addressed. This article presents an innovative approach that enhances the empirical likelihood-based information borrowing method by integrating missing-data techniques, ensuring robust data integration. We introduce a plug-in inverse probability weighting estimator to handle missingness in the primary analysis, demonstrating its equivalence to the standard joint estimator under mild conditions. To address potential bias from missing secondary outcomes, we propose a uniform mapping strategy, imputing incomplete secondary outcomes into a unified space. Extensive simulations highlight the effectiveness of our method, showing consistent, efficient, and robust estimators under various scenarios involving missing data and/or misspecified secondary models. Finally, we apply our proposal to the Uniform Data Set from the National Alzheimer’s Coordinating Center, exemplifying its practical application.

Using computational tools, bulk transcriptomics can be deconvoluted to estimate the abundance of constituent cell types. However, existing deconvolution methods are conditioned on the assumption that the whole study population is served by a single reference panel, ignoring person-to-person heterogeneity. Here, we present imply, a novel algorithm to deconvolute cell type proportions using personalized reference panels. Simulation studies demonstrate reduced bias compared with existing methods. Real data analyses on longitudinal consortia show disparities in cell type proportions are associated with several disease phenotypes in Type 1 diabetes and Parkinson’s disease. imply is available through the R/Bioconductor package ISLET at https://bioconductor.org/packages/ISLET/.

Recently, novel biotechnologies to quantify RNA modifications became an increasingly popular choice for researchers who study epitranscriptome. When studying RNA methylations such as N6-methyladenosine (m⁶A), researchers need to make several decisions in its experimental design, especially the sample size and a proper statistical power. Due to the complexity and high-throughput nature of m⁶A sequencing measurements, methods for power calculation and study design are still currently unavailable. In this work, we propose a statistical power assessment tool, magpie, for power calculation and experimental design for epitranscriptome studies using m⁶A sequencing data. Our simulation-based power assessment tool will borrow information from real pilot data, and inspect various influential factors including sample size, sequencing depth, effect size, and basal expression ranges. We integrate two modules in magpie: (i) a flexible and realistic simulator module to synthesize m⁶A sequencing data based on real data; and (ii) a power assessment module to examine a set of comprehensive evaluation metrics.

Real-world clinical samples are often admixtures of signal mosaics from multiple pure cell types. Using computational tools, bulk transcriptomics can be deconvoluted to solve for the abundance of constituent cell types. However, existing deconvolution methods are conditioned on the assumption that the whole study population is served by a single reference panel, which ignores person-to-person heterogeneity. Here we present imply , a novel algorithm to deconvolute cell type proportions using personalized reference panels. imply can borrow information across repeatedly measured samples for each subject, and obtain precise cell type proportion estimations. Simulation studies demonstrate reduced bias in cell type abundance estimation compared with existing methods. Real data analyses on large longitudinal consortia show more realistic deconvolution results that align with biological facts. Our results suggest that disparities in cell type proportions are associated with several disease phenotypes in type 1 diabetes and Parkin-son’s disease. Our proposed tool imply is available through the R/Bioconductor package ISLET at https://bioconductor.org/packages/ISLET/ .