Computational Methods for Single-cell DNA Methylome Analysis

Dissecting intercellular epigenetic differences is key to understanding tissue heterogeneity. Recent advances in single-cell DNA methylome profiling have presented opportunities to resolve this heterogeneity at the maximum resolution. While these advances enable us to explore frontiers of chromatin biology and better understand cell lineage relationships, they pose new challenges in data processing and interpretation. This review surveys the current state of computational tools developed for single-cell DNA methylome data analysis. We discuss critical components of single-cell DNA methylome data analysis, including data preprocessing, quality control, imputation, dimensionality reduction, cell clustering, supervised cell annotation, cell lineage reconstruction, gene activity scoring, and integration with transcriptome data. We also highlight unique aspects of single-cell DNA methylome data analysis and discuss how techniques common to other single-cell omics data analyses can be adapted to analyze DNA methylomes. Finally, we discuss existing challenges and opportunities for future development.

Aerosols over East and South Asia: Type Identification, Optical Properties, and Implications for Radiative Forcing

Identification of aerosol types has long been a difficult problem over East and South Asia due to various limitations. In this study, we use 2-dimensional (2-D) and multi-dimensional Mahalanobis distance (MD) clustering algorithms to identify aerosol characteristics based on the data from the Aerosol Robotic Network from March 1998 to February 2018 over the South and East Asian region (10°N~50°N, 70°E~135°E). The single scattering albedo (SSA), absorption Angstrom exponent (AAE), extinction Angstrom exponent (EAE), real index of refraction (RRI), and imaginary index of refraction (IRI) are utilized for classification of aerosols. Sub-regions with similar background conditions over East and South Asia are identified by hierarchical clustering algorithm to illustrate distinctive meteorological states in different areas. The East and South Asian aerosols are found to have distinct regional and seasonal features relating to the meteorological conditions, land cover, and industrial infrastructure. It is found that the proportions of dust aerosol are the highest in spring at the SACOL site and in summer at the sites near the Northern Indo-Gangetic Plain area. In spring, biomass-burning aerosols are dominant over the central Indo-China Peninsula area. The aerosol characteristics at coastal sites are also analyzed and compared with previous results. The 2-D clustering method is useful when limited aerosol parameters are available, but the results are highly dependent on the sets of parameters used for identification. Comparatively, the MD method, which considers multiple aerosol parameters, could provide more comprehensive classification of aerosol types. It is estimated that only about 50% of the data samples that are identifiable by the MD method could be classified by the 2-D methods, and a lot of undetermined data samples could be mis-classified by the 2-D methods. The aerosol radiative forcing (ARF) and the aerosol radiative forcing efficiency (ARFE) of various aerosol types at the top and the bottom of the atmosphere (TOA and BOA) are determined based on the MD aerosol classification. The dust aerosols are found to have the largest ARF at the TOA (−36 W/m2), followed by the urban/industrial aerosols and biomass-burning aerosols. The ARFE of biomass-burning aerosols at the BOA (−165 W/m2/AOD550nm) is the strongest among those of the other aerosol types. The comparison of the results by MD and 2-D methods shows that the differences in ARF and ARFE are generally within 10%. Our results indicate the importance of aerosol type classification in accurately attributing the radiative contributions of different aerosol components.

An Inception Convolutional Autoencoder Model for Chinese Healthcare Question Clustering

Healthcare question answering (HQA) system plays a vital role in encouraging patients to inquire for professional consultation. However, there are some challenging factors in learning and representing the question corpus of HQA datasets, such as high dimensionality, sparseness, noise, nonprofessional expression, etc. To address these issues, we propose an inception convolutional autoencoder model for Chinese healthcare question clustering (ICAHC). First, we select a set of kernels with different sizes using convolutional autoencoder networks to explore both the diversity and quality in the clustering ensemble. Thus, these kernels encourage to capture diverse representations. Second, we design four ensemble operators to merge representations based on whether they are independent, and input them into the encoder using different skip connections. Third, it maps features from the encoder into a lower-dimensional space, followed by clustering. We conduct comparative experiments against other clustering algorithms on a Chinese healthcare dataset. Experimental results show the effectiveness of ICAHC in discovering better clustering solutions. The results can be used in the prediction of patients' conditions and the development of an automatic HQA system.