Project description:Medullary thymic epithelial cells (mTECs) are critical for self-tolerance induction in T cells via promiscuous expression of tissue-specific antigens (TSAs), which are controlled by the transcriptional regulator, AIRE. Whereas AIRE-expressing (Aire+) mTECs undergo constant turnover in the adult thymus, mechanisms underlying differentiation of postnatal mTECs remain to be discovered. Integrative analysis of single-cell assays for transposase-accessible chromatin (scATAC-seq) and single-cell RNA sequencing (scRNA-seq) suggested the presence of proliferating mTECs with a specific chromatin structure, which express high levels of Aire and co-stimulatory molecules, CD80 (Aire+CD80hi). Proliferating Aire+CD80hi mTECs detected using Fucci technology express a minimal number of Aire-dependent TSAs and are converted into quiescent Aire+CD80hi mTECs expressing high levels of TSAs after a transit amplification. These data provide evidence for the existence of transit-amplifying Aire+mTEC precursors during the Aire+mTEC differentiation process of the postnatal thymus.
Project description:MotivationscATAC-seq has enabled chromatin accessibility landscape profiling at the single-cell level, providing opportunities for determining cell-type-specific regulation codes. However, high dimension, extreme sparsity, and large scale of scATAC-seq data have posed great challenges to cell-type identification. Thus, there has been a growing interest in leveraging the well-annotated scRNA-seq data to help annotate scATAC-seq data. However, substantial computational obstacles remain to transfer information from scRNA-seq to scATAC-seq, especially for their heterogeneous features.ResultsWe propose a new transfer learning method, scNCL, which utilizes prior knowledge and contrastive learning to tackle the problem of heterogeneous features. Briefly, scNCL transforms scATAC-seq features into gene activity matrix based on prior knowledge. Since feature transformation can cause information loss, scNCL introduces neighborhood contrastive learning to preserve the neighborhood structure of scATAC-seq cells in raw feature space. To learn transferable latent features, scNCL uses a feature projection loss and an alignment loss to harmonize embeddings between scRNA-seq and scATAC-seq. Experiments on various datasets demonstrated that scNCL not only realizes accurate and robust label transfer for common types, but also achieves reliable detection of novel types. scNCL is also computationally efficient and scalable to million-scale datasets. Moreover, we prove scNCL can help refine cell-type annotations in existing scATAC-seq atlases.Availability and implementationThe source code and data used in this paper can be found in https://github.com/CSUBioGroup/scNCL-release.
Project description:It is a challenging task to integrate scRNA-seq and scATAC-seq data obtained from different batches. Existing methods tend to use a pre-defined gene activity matrix to convert the scATAC-seq data into scRNA-seq data. The pre-defined gene activity matrix is often of low quality and does not reflect the dataset-specific relationship between the two data modalities. We propose scDART, a deep learning framework that integrates scRNA-seq and scATAC-seq data and learns cross-modalities relationships simultaneously. Specifically, the design of scDART allows it to preserve cell trajectories in continuous cell populations and can be applied to trajectory inference on integrated data.
Project description:Powerful next generation sequencing techniques offer robust and comprehensive analysis to investigate how retinal gene regulatory networks function during development and in disease states. Single-cell RNA sequencing allows us to comprehensively profile gene expression changes observed in retinal development and disease at a cellular level, while single-cell ATAC-Seq allows analysis of chromatin accessibility and transcription factor binding to be profiled at similar resolution. Here the use of these techniques in the developing retina is described, and MULTI-Seq is demonstrated, where individual samples are labeled with a modified oligonucleotide-lipid complex, enabling researchers to both increase the scope of individual experiments and substantially reduce costs.
