Page 54 - 2019_05-HaematologicaMondo-web

Z. Wu et al.
A
BCD
E
F
Figure 4. Long noncoding RNA expression exhibited cell lineage-specific patterns. (A) The same t-SNE plot as in Figure 3A, highlighted single cells with a CD38 expression level (left); cell types were assigned to single cells using messenger RNA (mRNA) expression information, followed by differentiation tree reconstruction using a pseudotime ordering method (middle); single cells are colored according to their corresponding cell types, and gray circles indicate clustering of the same cell type (right). (B) Variance of long noncoding RNA (lncRNA) versus. mRNA expression among different lineages. x axis, Log(TPM+1); y axis, variance. (C) JScore to assess lineage specificity of lncRNA and mRNA expression. x axis, JScore; y axis, cumulative distribution function (CDF). (D) Percentages of mRNAs and lncRNAs defined (top) or preferentially expressed (bottom) in various numbers of cell types. HSC: hematopoietic stem cell; MLP: multilymphoid progenitor; MEP: megakary- ocyte-erythroid progenitor; GMP: granulocyte-monocyte progenitor, ProB: pro-B cell; ETP: earliest thymic progenitor. (E) Heatmaps of mRNA (left) and lncRNA (right) expression in different lineages. (F) Expression of a group of lineage-specific lncRNAs for HSC/MLP, MEP, ProB, and ETP along the differentiation tree, measured by single cell RNA sequencing (scRNA-seq) (left) and quantitative reverse transcriptase polymerase chain reaction (RT-PCR) analysis (right). Expression (shown as a mean expression level) is presented as a relative quantity in one lineage relative to expression in all the others.
900
haematologica | 2019; 104(5)