A systematic assessment of cell type deconvolution algorithms for DNA methylation data

A systematic evaluation of cell-type-specific differential methylation analysis in bulk tissue

We conducted a systematic assessment of computational models-CellDMC, TCA, HIRE, TOAST, and CeDAR-for detecting cell-type-specific differential methylation CpGs in bulk methylation data profiled using the Illumina DNA Methylation BeadArrays. This assessment was performed through simulations and case studies involving two epigenome-wide association studies (EWAS) on rheumatoid arthritis and major depressive disorder. Our evaluation provided insights into the strengths and limitations of each model. The results revealed that the models varied in performance across different metrics, sample sizes, and computational efficiency. Additionally, we proposed integrating the results from these models using the minimum p-value ($minpv$) and average p-value ($avepv$) approaches. Our findings demonstrated that these aggregation methods significantly improved performance in identifying cell-type-specific differential methylation CpGs.

Role of epigenetic mechanisms in the pathogenesis of chronic respiratory diseases and response to inhaled exposures: From basic concepts to clinical applications

Epigenetic modifications are chemical groups in our DNA (and chromatin) that determine which genes are active and which are shut off. Importantly, they integrate environmental signals to direct cellular function. Upon chronic environmental exposures, the epigenetic signature of lung cells gets altered, triggering aberrant gene expression programs that can lead to the development of chronic lung diseases. In addition to driving disease, epigenetic marks can serve as attractive lung disease biomarkers, due to early onset, disease specificity, and stability, warranting the need for more epigenetic research in the lung field. Despite substantial progress in mapping epigenetic alterations (mostly DNA methylation) in chronic lung diseases, the molecular mechanisms leading to their establishment are largely unknown. This review is meant as a guide for clinicians and lung researchers interested in epigenetic regulation with a focus on DNA methylation. It provides a short introduction to the main epigenetic mechanisms (DNA methylation, histone modifications and non-coding RNA) and the machinery responsible for their establishment and removal. It presents examples of epigenetic dysregulation across a spectrum of chronic lung diseases and discusses the current state of epigenetic therapies. Finally, it introduces the concept of epigenetic editing, an exciting novel approach to dissecting the functional role of epigenetic modifications. The promise of this emerging technology for the functional study of epigenetic mechanisms in cells and its potential future use in the clinic is further discussed.

Methods in DNA methylation array dataset analysis: A review

Understanding the intricate relationships between gene expression levels and epigenetic modifications in a genome is crucial to comprehending the pathogenic mechanisms of many diseases. With the advancement of DNA Methylome Profiling techniques, the emphasis on identifying Differentially Methylated Regions (DMRs/DMGs) has become crucial for biomarker discovery, offering new insights into the etiology of illnesses. This review surveys the current state of computational tools/algorithms for the analysis of microarray-based DNA methylation profiling datasets, focusing on key concepts underlying the diagnostic/prognostic CpG site extraction. It addresses methodological frameworks, algorithms, and pipelines employed by various authors, serving as a roadmap to address challenges and understand changing trends in the methodologies for analyzing array-based DNA methylation profiling datasets derived from diseased genomes. Additionally, it highlights the importance of integrating gene expression and methylation datasets for accurate biomarker identification, explores prognostic prediction models, and discusses molecular subtyping for disease classification. The review also emphasizes the contributions of machine learning, neural networks, and data mining to enhance diagnostic workflow development, thereby improving accuracy, precision, and robustness.

Targeting the hypothalamus for modeling age-related DNA methylation and developing OXT-GnRH combinational therapy against Alzheimer's disease-like pathologies in male mouse model

The hypothalamus plays an important role in aging, but it remains unclear regarding the underlying epigenetics and whether this hypothalamic basis can help address aging-related diseases. Here, by comparing mouse hypothalamus with two other limbic system components, we show that the hypothalamus is characterized by distinctively high-level DNA methylation during young age and by the distinct dynamics of DNA methylation and demethylation when approaching middle age. On the other hand, age-related DNA methylation in these limbic system components commonly and sensitively applies to genes in hypothalamic regulatory pathways, notably oxytocin (OXT) and gonadotropin-releasing hormone (GnRH) pathways. Middle age is associated with transcriptional declines of genes which encode OXT, GnRH and signaling components, which similarly occur in an Alzheimer's disease (AD)-like model. Therapeutically, OXT-GnRH combination is substantially more effective than individual peptides in treating AD-like disorders in male 5×FAD model. In conclusion, the hypothalamus is important for modeling age-related DNA methylation and developing hypothalamic strategies to combat AD.

Neural-net-based cell deconvolution from DNA methylation reveals tumor microenvironment associated with cancer prognosis

DNA methylation is a pivotal epigenetic modification that defines cellular identity. While cell deconvolution utilizing this information is considered useful for clinical practice, current methods for deconvolution are limited in their accuracy and resolution. In this study, we collected DNA methylation data from 945 human samples derived from various tissues and tumor-infiltrating immune cells and trained a neural network model with them. The model, termed MEnet, predicted abundance of cell population together with the detailed immune cell status from bulk DNA methylation data, and showed consistency to those of flow cytometry and histochemistry. MEnet was superior to the existing methods in the accuracy, speed, and detectable cell diversity, and could be applicable for peripheral blood, tumors, cell-free DNA, and formalin-fixed paraffin-embedded sections. Furthermore, by applying MEnet to 72 intrahepatic cholangiocarcinoma samples, we identified immune cell profiles associated with cancer prognosis. We believe that cell deconvolution by MEnet has the potential for use in clinical settings.

Benchmarking of methods for DNA methylome deconvolution

Defining the number and abundance of different cell types in tissues is important for understanding disease mechanisms as well as for diagnostic and prognostic purposes. Typically, this is achieved by immunohistological analyses, cell sorting, or single-cell RNA-sequencing. Alternatively, cell-specific DNA methylome information can be leveraged to deconvolve cell fractions from a bulk DNA mixture. However, comprehensive benchmarking of deconvolution methods and modalities was not yet performed. Here we evaluate 16 deconvolution algorithms, developed either specifically for DNA methylome data or more generically. We assess the performance of these algorithms, and the effect of normalization methods, while modeling variables that impact deconvolution performance, including cell abundance, cell type similarity, reference panel size, method for methylome profiling (array or sequencing), and technical variation. We observe differences in algorithm performance depending on each these variables, emphasizing the need for tailoring deconvolution analyses. The complexity of the reference, marker selection method, number of marker loci and, for sequencing-based assays, sequencing depth have a marked influence on performance. By developing handles to select the optimal analysis configuration, we provide a valuable source of information for studies aiming to deconvolve array- or sequencing-based methylation data.

Validating reference-based algorithms to determine cell-type heterogeneity in ovarian cancer DNA methylation studies

Information about cell composition in tissue samples is crucial for biomarker discovery and prognosis. Specifically, cancer tissue samples present challenges in deconvolution studies due to mutations and genetic rearrangements. Here, we optimized a robust, DNA methylation-based protocol, to be used for deconvolution of ovarian cancer samples. We compared several state-of-the-art methods (HEpiDISH, MethylCIBERSORT and ARIC) and validated the proposed protocol in an in-silico mixture and in an external dataset containing samples from ovarian cancer patients and controls. The deconvolution protocol we eventually implemented is based on MethylCIBERSORT. Comparing deconvolution methods, we paid close attention to the role of a reference panel. We postulate that a possibly high number of samples (in our case: 247) should be used when building a reference panel to ensure robustness and to compensate for biological and technical variation between samples. Subsequently, we tested the performance of the validated protocol in our own study cohort, consisting of 72 patients with malignant and benign ovarian disease as well as in five external cohorts. In conclusion, we refined and validated a reference-based algorithm to determine cell type composition of ovarian cancer tissue samples to be used in cancer biology studies in larger cohorts.

Computational deconvolution of DNA methylation data from mixed DNA samples

In this review, we provide a comprehensive overview of the different computational tools that have been published for the deconvolution of bulk DNA methylation (DNAm) data. Here, deconvolution refers to the estimation of cell-type proportions that constitute a mixed sample. The paper reviews and compares 25 deconvolution methods (supervised, unsupervised or hybrid) developed between 2012 and 2023 and compares the strengths and limitations of each approach. Moreover, in this study, we describe the impact of the platform used for the generation of methylation data (including microarrays and sequencing), the applied data pre-processing steps and the used reference dataset on the deconvolution performance. Next to reference-based methods, we also examine methods that require only partial reference datasets or require no reference set at all. In this review, we provide guidelines for the use of specific methods dependent on the DNA methylation data type and data availability.

Cell type signatures in cell-free DNA fragmentation profiles reveal disease biology

Circulating cell-free DNA (cfDNA) fragments have characteristics that are specific to the cell types that release them. Current methods for cfDNA deconvolution typically use disease tailored marker selection in a limited number of bulk tissues or cell lines. Here, we utilize single cell transcriptome data as a comprehensive cellular reference set for disease-agnostic cfDNA cell-of-origin analysis. We correlate cfDNA-inferred nucleosome spacing with gene expression to rank the relative contribution of over 490 cell types to plasma cfDNA. In 744 healthy individuals and patients, we uncover cell type signatures in support of emerging disease paradigms in oncology and prenatal care. We train predictive models that can differentiate patients with colorectal cancer (84.7%), early-stage breast cancer (90.1%), multiple myeloma (AUC 95.0%), and preeclampsia (88.3%) from matched controls. Importantly, our approach performs well in ultra-low coverage cfDNA datasets and can be readily transferred to diverse clinical settings for the expansion of liquid biopsy.

Challenges and perspectives in computational deconvolution of genomics data

Deciphering cell-type heterogeneity is crucial for systematically understanding tissue homeostasis and its dysregulation in diseases. Computational deconvolution is an efficient approach for estimating cell-type abundances from a variety of omics data. Despite substantial methodological progress in computational deconvolution in recent years, challenges are still outstanding. Here we enlist four important challenges related to computational deconvolution: the quality of the reference data, generation of ground truth data, limitations of computational methodologies, and benchmarking design and implementation. Finally, we make recommendations on reference data generation, new directions of computational methodologies, and strategies to promote rigorous benchmarking.