Enrichment of imprinted genes in somatic stem cell populations from published gene expression studies.

Datasets from the Gene Expression Omnibus (GEO) that included stem cells together with a comparator population were used to examine enrichment for the IGN using Gene Set Enrichment Analysis (GSEA) with a custom imprinted-gene gene set. As originally defined [16], the IGN includes 16 genes, but for this analysis we also included additional imprinted genes that were specific to the LT-HSCs in our study [11]. This combined IGN-HSC list includes Asb4, Cdkn1c, Dcn, Dlk1, Gatm, Gnas, Grb10, Gtl2, H19, Igf2, Igf2r, Impact, Mest, Ndn, Nnat, Peg3, Peg10, Peg12, Plagl1, Ppp1r9a, Sgce, Slc22a3, Slc22a18, and Slc38a4.
(a)The nominal P-value for all tests indicating 0.000 is <0.0001 (1000 permutations were used in each test). HSC, hematopoietic stem cells; MPP, multipotential progenitors; NPC, neural progenitor cells; ESC, embryonic stem cells; EB, embryoid bodies; AML, acute myeloid leukemia; FDR, false discovery rate; GEO, gene expression omnibus; PMID, pubmed ID; ns, not-significant.

本研究采用了包含干细胞及其对照群体的基因表达综合数据库（Gene Expression Omnibus, GEO）数据集，通过自定义印记基因集结合基因集富集分析（Gene Set Enrichment Analysis, GSEA），对印记基因网络（Imprinted Gene Network, IGN）的富集情况进行检测。
如最初的定义所述[16]，该印记基因网络（IGN）原本包含16个基因，但为开展本次分析，我们还纳入了本研究中特异性针对长期造血干细胞（long-term hematopoietic stem cells, LT-HSCs）的额外印记基因[11]。本次整合得到的IGN-HSC基因集包含以下基因：Asb4、Cdkn1c、Dcn、Dlk1、Gatm、Gnas、Grb10、Gtl2、H19、Igf2、Igf2r、Impact、Mest、Ndn、Nnat、Peg3、Peg10、Peg12、Plagl1、Ppp1r9a、Sgce、Slc22a3、Slc22a18及Slc38a4。
(a) 所有检验的名义P值均显示为0.000，即小于0.0001（每项检验均采用1000次置换检验）。以下为各缩写的详细释义：造血干细胞（hematopoietic stem cells, HSC）、多能祖细胞（multipotential progenitors, MPP）、神经祖细胞（neural progenitor cells, NPC）、胚胎干细胞（embryonic stem cells, ESC）、拟胚体（embryoid bodies, EB）、急性髓系白血病（acute myeloid leukemia, AML）、错误发现率（false discovery rate, FDR）、基因表达综合数据库（Gene Expression Omnibus, GEO）、PubMed编号（PubMed ID, PMID）、无统计学显著性（not-significant, ns）。