Frequently Asked Questions

Users are welcome to send us additional questions regarding our data resource. We will add common questions and their answers to our list.

What platform(s) were used in the study?
The Affymetrix Mouse 430 2.0. Please go to Affymetrix for technical documentation on this platform.
How were the samples pre-processed?
Probes were summarized using Robust Multichip Average (RMA) and chips were normalized using quantile normalization as implemented in R/BioConductor. ComBat was used to correct for batch effects. Only the 13,530 genes detected as 'Present' in at least one biological group were used in finding sets of co-regulated genes, in differential expression, and in network analysis and only these genes are included as part of this online resource.
How were the sets of co-regulated genes (modules) identified?
The Weighted Gene Co-expression Network Analysis (WGCNA) algorithm was used to find sets of positively co-regulated genes.
What is a module profile?
The expression profiles of the module genes were simplified to one profile per module. This was done by first standardizing each gene profile, then the module value for each sample was computed as the median of the standardized gene expression values for that module.
How were putative transcriptional regulators of HSC development identified?
The Context Likelihood of Relatedness (CLR) algorithm was applied to module profiles and standardized gene expression of transcriptional regulators.
The website says that it cannot find my gene of interest. Why not?
There are two reasons that you may not be able to find your gene of interest. The first is that the gene symbol that you entered does not match the official MGI symbols in the StemSite database. If this is the case, you should receive a message stating that this is the case. You can find the official gene symbol at MGI or, for lists of gene identifiers, you can use DAVID for conversion to gene symbols.
The second reason that your gene may not be found is because it is not expressed at a detectable level in the samples profiled. Again, you should receive a message stating that this is the case.
What are ESC-HSC cells?
Please see the following references: Kyba et al, Cell, 2002; Wang et al, PNAS, 2005; and McKinney-Freeman et al, Blood, 2009.
How were embryoid bodies generated for these experiments?
Embryoid bodies (EBs) were generated as described in the following references from the laboratory of George Daley: Kyba et al, Cell, 2002; Wang et al, PNAS, 2005; McKinney-Freeman et al, Blood, 2008; McKinney-Freeman et al, Blood, 2009;
The following abbreviations are used throughout this site:
AGM: aorta-gonads-mesonephros
CLR: Context Likelihood of relatedness
DAVID: Database for Annotation, Visualization, and Integrated Discovery
E: Embryonic day (i.e. E9 = embryonic day 9)
EB: embryoid body
ESC: embryonic stem cell
FL: Fetal Liver
FDR: False Discovery Rate
HSC: hematopoietic stem cell
MGI: Mouse Genomic Informatics
NCBI: National Center for Biotechnology Information
Pla: placenta
WBM: whole bone marrow
WGCNA: Weighted Gene Co-expression Network Analysis
YS: yolk sac
What are the P values in the module searching results?
We perform the two-tail Fisher Exact probability test to test the statistical significance of the input genes list overlapping with the module gene list. Fisher Exact test is a non-parametric statistical test based upon the hypergeometric probability used to determine if there are nonrandom associations between the two categorical variables. In our case it is calculating between the user input gene list and the genes in different modules. Reference: Fisher, R.A. (1954). Statistical Methods for Research Workers. Oliver and Boyd. ISBN 0050021702.
How can I cite this data?
McKinney-Freeman S*, Cahan P*, Li H*, Lacadie SA, Huang HT, Curran M, Loewer S, Naveiras O, Kathrein KL, Konantz M, Langdon EM, Lengerke C, Zon LI, Collins JJ, Daley GQ. The transcriptional landscape of hematopoietic stem cell ontogeny. Cell Stem Cell. 2012 Nov 2;11(5):701-14. doi: 10.1016/j.stem.2012.07.018. PubMed PMID: 23122293. * Denotes equal contribution