Ph.D. in Biomedical Engineering, Mayo Clinic, 2005
B.S. in Chemical Physics, University of Science and Technology of China, 2000



2011-2013: Postdoctoral fellow at the Wang Genomics Lab (joint with Dr. Wange Lu)


Current Position

Supervisor of Bioinformatics, Picower Institute for Learning and Memory, MIT


Research Area

higher-order chromosomal structure and interactions, epigenetic and transcriptome analysis of meningiomas



Higher-order chromosomal structure and interaction

The genes POU5F1 and SOX2 are critical for pluripotency and reprogramming, yet the chromosomal organization around these genes remains poorly understood. We assayed long-range chromosomal interactions on putative enhancers of POU5F1 and SOX2 genes in human embryonic stem cells (hESCs) using 4C-Seq technique. We discovered that their frequent interacting regions mainly overlap with early DNA replication domains. The interactomes are associated with active histone marks and enriched with 5-hydroxymethylcytosine sites. In hESCs, genes within the interactomes have elevated expression. Additionally, some genes associated with the POU5F1 enhancer contribute to pluripotency. Binding sites for multiple DNA binding proteins, including ATF3, CTCF, GABPA, JUND, NANOG, RAD21 and YY1, are enriched in both interactomes. The RARG locus, frequently interacting with the POU5F1 locus, has abundant RAD21 binding sites co-localized with other protein binding sites. Thus the interactomes of these two pluripotency genes could be an important part of the regulatory network in hESCs.

In a different study, we compared the performance of enzyme-based and sonication-based methods for 4C-Seq:

Circular chromosome conformation capture, when coupled with next-generation sequencing (4C-Seq), can be used to identify genome-wide interaction of a given locus (a "bait" sequence) with all of its interacting partners. Conventional 4C approaches used restriction enzyme digestion to fragment chromatin, and recently sonication approach was also applied for this purpose. However, bioinformatics pipelines for analyzing sonication-based 4C-Seq data are not well developed. In addition, data consistency as well as similarity between the two methods has not been explored previously. Here we present a comparative analysis of 4C-Seq data generated by both methods, using an enhancer element of Pou5f1 gene in mouse embryonic stem (ES) cells.

From biological replicates, we found good correlation (r>0.6) for inter-chromosomal interactions identified in either enzyme or sonication method. Compared to enzyme approach, sonication method generated less distal intra-chromosomal interactions, possibly due to the difference in chromatin fragmentation. From all mapped interactions, we further applied statistical models to identify enriched interacting regions. Interestingly, data generated from the two methods showed 30% overlap of the reproducible interacting regions. The interacting sites in the reproducible regions from both methods are similarly enriched with active histone marks. In addition, the interacting sites identified from sonication-based data are enriched with ChIP-Seq signals of transcription factors Oct4, Klf4, Esrrb, Tcfcp2i1, and Zfx that are critical for reprogramming and pluripotency.

Both enzyme-based and sonication-based 4C-Seq methods are valuable tools to explore long-range chromosomal interactions. Due to the nature of sonication-based method, correlation analysis of the 4C interactions with transcription factor binding should be more straightforward.


Epigenetic and transcriptome analysis of meningiomas 

Comparison of epigenetic marks at HOXA and MAL2 gene loci

Meningiomas are central nervous system tumors that originate from the meningeal coverings of the brain and spinal cord. Most meningiomas are pathologically benign or atypical, but 3-5% display malignant features. Despite previous studies on benign and atypical meningiomas, the key molecular pathways involved in malignant transformation remain to be determined, as does the extent of epigenetic alteration in malignant meningiomas.

In this study, we explored the landscape of DNA methylation in ten benign, five atypical and four malignant meningiomas. Compared to the benign tumors, the atypical and malignant meningiomas demonstrate increased global DNA hypomethylation. Clustering analysis readily separates malignant from atypical and benign tumors, implicating that DNA methylation patterns may serve as diagnostic biomarkers for malignancy. Genes with hypermethylated CpG islands in malignant meningiomas (such as HOXA6 and HOXA9) tend to coincide with the binding sites of polycomb repressive complexes (PRC) in early developmental stages. Most genes with hypermethylated CpG islands at promoters are suppressed in malignant and benign meningiomas, suggesting the switching of gene silencing machinery from PRC binding to DNA methylation in malignant meningiomas. One exception is the MAL2 gene that is highly expressed in benign group and silenced in malignant group, representing de novo gene silencing induced by DNA methylation.

In summary, our results suggest that malignant meningiomas have distinct DNA methylation patterns compared to their benign and atypical counterparts, and that the differentially methylated genes may serve as diagnostic biomarkers or candidate causal genes for malignant transformation.




  1. Gao F, Lin E, Feng Y, Mack WJ, Shen Y, Wang K. Characterizing immunoglobulin repertoire from whole blood by a personal genome sequencer. PLoS ONE, 8:e75294, 2013
  2. Gao F, Ling C, Shi L, Commins D, Zada G, Mack W, Wang K. Inversion-mediated gene fusion involving NAB2-STAT6 in an unusual malignant meningioma. British Journal of Cancer, 109:1051-1055, 2013
  3. Chang X, Shi L, Gao F, Russin J, Zeng L, He S, Chen TC, Giannotta SL, Weisenberger DJ, Zada G, Wang K*, Mack WJ*. Genomic and transcriptomic analysis reveals an oncogenic functional module in meningiomas. Neurosurgical Focus, 35:e3, 2013
  4. Wei Z, Huang D, Gao F, Chang WH, Woojin A, Coetzee GA, Wang K, Lu W. The biological implications and regulatory mechanisms of long-range chromosomal interactions. Journal of Biological Chemistry, 288:22369-22377, 2013
  5. Wei Z, Gao F, Kim S, Yang H, Wang K, Lu W. Klf4 Organizes Long-Range Chromosomal Interactions with the Oct4 Locus in Reprogramming and Pluripotency. Cell Stem Cell, 13:36-47, 2013
  6. Gao F, Wei Z, Lu W, Wang K. Comparative analysis of 4C-Seq data generated from enzyme-based and sonication-based methods. BMC Genomics, 14:345, 2013
  7. Gao F, Wei Z, Woo A, Wang K*, Lu W*. The interactomes of POU5F1 and SOX2 enhancers in human embryonic stem cells. Scientific Reports, 3:1588, 2013 
  8. Gao F, Shi L, Russin J, Zeng L, Chang X, He S, Chen TC, Giannotta SL, Weisenberger DJ, Zada G, Mack WJ, Wang K. DNA methylation in the malignant transformation of meningiomas. PLoS ONE, 8:e54114, 2013
  9. Wang K*, Edmondson AC, Li M, Gao F, Qasim AN, Devaney JM, Burnett MS, Waterworth DM, Mooser V, Grant SF, Epstein SE, Reilly MP, Hakonarson H, Rader DJ. Pathway-wide association study implicates multiple sterol transport and metabolism genes in HDL cholesterol regulation. Frontiers in Genetics, 2:41, 2011