The Stem Cell Discovery Engine (SCDE) is an integrated platform that allows users to consistently describe, share and compare cancer and tissue stem cell data. It is made up of an online database of curated experiments coupled to a customized instance of the Galaxy analysis engine with tools for gene list manipulation and molecular profile comparisons. The SCDE currently contains more than 50 stem cell-related experiments. Each has been manually curated and encoded using the ISA-Tab standard to ensure the quality of the data and its annotation.
The use of open source tools and community-driven standards was motivated by the desire to establish the SCDE as a community resource that encourages contributions of tools and new data sets from other stem cell researchers. The ISA-Tab framework is gaining support as a standard for scalable data capture and annotation, and makes it possible for SCDE to accommodate diverse data types. The Galaxy development community is growing rapidly and as a result, new methods are quickly being integrated into this framework, many of which will be applicable to stem cell data analysis. We are actively collaborating with the Galaxy team as well as developing our own tools for Galaxy, and will continue to align with this community resource as it is likely to yield benefits as we scale up and acquire new data and new data types.
CHB has supported the development of the SCDE with funding from the “Harvard Stem Cell Institute”: and an NIH GO grant on Comprehensive Phenotypic Comparison of Normal and Cancer Stem Cells. Shannan Ho Sui, the lead author on the the NAR paper would love to hear your feedback.
Ho Sui, S. J., Begley, K., Reilly, D., Chapman, B., McGovern, R., Rocca-Sera, P., Maguire, E., et al. (2011). The Stem Cell Discovery Engine: an integrated repository and analysis system for cancer stem cell comparisons. Nucleic acids research
MicroRNAs (miRNAs) are small RNAs that regulate gene expression by binding to mRNAs bearing a partially complementary sequence. miRNAs decrease the stability or translation of mRNA targets, leading to reduced protein expression. Understanding the biological function of a miRNA requires identifying its targets, and current target prediction systems yield a high false-positive rate.
Because we have a background in working with complex biological network relationships, CHB was invited to support a collaborative effort led by Ashish Lal from the Lieberman lab at the Immune Disease Institute, including researchers from Harvard Medical School, Harvard School of Public Health, Children’s Hospital, NCI and Memorial Sloan Kettering Cancer Center.
As part of the project just published in PLoS Genetics Ashish developed a sensitive and specific biochemical method to identify candidate microRNA targets that are enriched by pull-down with a tagged, transfected microRNA mimic. The method was used to isolate mRNAs pulled down with a transfected, biotinylated mir-34a, a tumor suppressor gene inhibiting cell proliferation, in K562 and HCT116 cancer cell lines.
One of the key challenges involved in the data analysis was to identify common biological functionality and interactions within the hundreds of identified target genes which we tackled by integrating gene-gene interaction information from KEGG, WikiPathways and GeneGo’s MetaCore.
Transcripts pulled down with miR-34a were highly enriched for roles in growth factor signaling and cell cycle progression, forming a dense network regulating multiple signal transduction pathways involved in the cell proliferative response.
Congratulations to Ashish and the whole team!
Endotoxin is a poorly recognized environmental exposure that has been associated with COPD development in both human cohorts and animal models. Several studies have used gene expression arrays to obtain pulmonary expression signatures in mouse models of chronic inhaled endotoxin. However, due to differences between platforms, study design and laboratory effects direct comparison of gene signatures has proven difficult.
We are supporting Dr Lai In collaboration with the groups of Dr Christiani and Dr Baron with the functional integration of gene expression data obtained from the different studies by focusing on shared pathways and processed instead.
Initial results show consistent functional enrichment between studies, but at different levels of significance. A general increase in significance of inflammatory pathways with extended enodoxin exposure was noticed across several studies. We are exploring the underlying pathiophysiology and its impact on endotoxin-related chronic obstructive lung disease.