Link to bioRxiv paper: http://biorxiv.org/cgi/content/short/2020.09.28.316901v1?rss=1 Authors: Shi, S. Abstract: We proposed a method for data integration and cell type annotation (Dincta) of single cell transcriptomes in a unify framework. The Dincta can handle three cases. In the first case, the data has been annotated the cell type for all cells, Dincta can integrate the the data into a common low dimension embedding space such that cells with different cell types separate while cells from the different batches but in the same cell type cluster together. In the second case, the data was only annotated for part of cells, such as one sample, Dincta can integrate the data into a common low dimension embedding space such that cells with different cell types separate while cells from the different batches but in the same cell type cluster together. Moreover, it can infer the known or novel cell type of the cells with unknown cell type initially. In the third case, there are no cell type information of cells, we can run Dincta in an unsupervised way. It can infer the number of new cell types and annotate the cells into its correspond cell type, and do data integration keeping cells from different cell type separate while removing the batch effects to mix cells in the same cell type. Dincta is simple, accurate and efficient to integrate data, which keeps the cell type information preserved while removes the batch effects, and infers the known or novel cell types of cells. Copy rights belong to original authors. Visit the link for more info