Introduction Pan-cancer study can uncover most of cell- and tissue- specific genomic loci and regions with underlying biological functions (Kristensen et al., 2014; Leiserson et al., 2015; The Cancer Genome Atlas Research Network et al., 2013; Witte et al., 2014). Meanwhile, it provides meaningful insights from the genome-wide interro- gation of cross-cell analysis and annotation. While till now, to biomedical researchers and clinicians, there is no systematic reference source of functional association between DNA methylation and transcrip- tional regulation for wet-lab experiment design and post-experiment validation. Thus, this is an imperative for most biologists and biomedical researchers to improve their research outcomes and efficiency (Bock and Lengauer, 2008; Roadmap Epigenomics Consor- tium et al., 2015). Here, we utilized our online curated reference source for DNA Methylation Annotation Knowledgebase (DMAK) and implemented the cross-cell analysis in pan -cancer study. The knowledgebase provides multiple read-to-use analysis results and annotation information for the pan-cancer interrogation and cross-validation. We deposited the curated information knowledgebase and related analysis results on GitHub for direct download and usage for free. Proper citation is suggested for any usage, possible reanalysis or refinement. A Journal of the Bangladesh Pharmacological Society (BDPS) Bangladesh J Pharmacol 2016; 11: S154-S160 Journal homepage: www.banglajol.info Abstracted/indexed in Academic Search Complete, Asia Journals Online, Bangladesh Journals Online, Biological Abstracts, BIOSIS Previews, CAB Abstracts, Current Abstracts, Directory of Open Access Journals, EMBASE/Excerpta Medica, Google Scholar, HINARI (WHO), International Pharmaceutical Abstracts, Open J-gate, Science Citation Index Expanded, SCOPUS and Social Sciences Citation Index; ISSN: 1991-0088 Abstract Pan-cancer study can uncover cell- and tissue-specific genomic loci and regions with underlying biological functions, as one of fundamental procedures toward precision medicine. We utilized the online curated resource of DNA methylation annotation knowledgebase, to implement the cross-cell interrogation of pan-cancer study of breast cancer. The study revealed genome-wide differentially-methylated loci and regions by the reduced representation bisulfite sequencing profiling. The knowledgebase contains three level of curated information across multiple cancer and normal cells from the ENCODE Consortium. The reference base covers all identified differentially-methylation CpG sites and regions of interest, further annotated gene information, together with tumor suppressor gene and methylation level. Lastly, it includes the inferred functional association network and related Gene Ontology analysis results based on all the tumor suppressor genes identified from the differentially-methylated regions of interest. Our knowledgebase and analysis results provide a thorough reference source for biomedical researchers and clinicians. The cross-cell analysis results are deposited at: http://github.com/gladex/DMAK. Article Info Received: 12 January 2016 Accepted: 27 February 2016 Available Online: 11 March 2016 DOI: 10.3329/bjp.v11iS1.26851 Cite this article: Tang B, Zhu W, Wu C. Cross-cell DNA methylation annotation and analysis for pan-cancer study. Bangladesh J Pharmacol. 2016; 11: S154-S160. Cross-cell DNA methylation annotation and analysis for pan-cancer study Binhua Tang1,2,3, Weiliang Zhu4 and Changping Wu4 1Epigenetics and Function Group, College of the Internet of Things, Hohai University, Jiangsu 213 022, China; 2School of Public Health, Shanghai Jiao Tong University, Shanghai 200 225, China; 3CBMI, Harvard Medical School, Boston, MA 02 115, USA; 4The Third Affiliated Hospital, Soochow University, Jiangsu 213 003, China. http://www.bioxbio.com/if/html/BANGL-J-PHARMACOL.html http://www.asiajol.info/index.php/record/view/50672 Structure and Purpose of DMAK DMAK contains three levels of curated information across multiple cell types from ENCODE Consortium portal (de Souza, 2012; Pennisi, 2012; Tang and Wang, 2015; The ENCODE Project Consortium, 2012). The cell types investigated as below include breast cancer (T- 47D and MCF-7), cervical cancer (HeLa-S3), endo- metrial cancer (ECC-1), blood cancer (GM12878, GM12891, GM12892, HL-60 and K562), brain cancer (SK -N-MC, SK-N-SH, SK-N-SH_RA, PFSK-1 and U87), liver cancer (HepG2), colon cancer (HCT-116), pancreas cancer (PANC-1), lung cancer (A549), and human embryonic stem cell (H1-hESC). As depicted in Figure 1, the first level of DMAK was the curation of raw data sources from ENCODE Consor- tium portal; for the study case in our work, we emphasized on cross-cell DNA methylation profiling information for detecting differentially-methylated fea- tures and patterns within breast cancer T-47D cell type. This level content includes the summary for the analysis procedure and fundamental functions as discussed in the following sections. The second level mainly focuses on integrative analysis on the curated DNA methylation data in RRBS format (Blattler et al., 2014; Ziller et al., 2013), we implemented function annotation for methylated CpG sites, identi- fied differentially-methylated regions (DMR), and classified the hyper- and hypo-methylated regions or differential DMR candidates (Kemp Christopher et al., 2014). The detailed analysis procedure and results are given in the following section. The third level analysis mainly includes the visualiza- tion and function analysis for the annotated results, which include the functional association network for tumor suppressor genes identified from the hyper- or hypo-DMRs detected from the above analysis. We curated information and constructed the compre- hensive knowledgebase using data sources mainly from ENCODE Consortium portal, together with other commonly-used tools, and the self-compiled scripts and programs. Annotation and Analysis Procedure in DMAK This section mainly discusses the functions and analysis procedure for DMAK, which covers fundamental functions of DMAK reference source, listed as, 1) Statistical information detected for sequencing read coverage, number of Cs and Ts of for the 688,445 CpG sites across all cancer and normal cell lines listed above. For consistence, all DNA methylation data sets from ENCODE Consortium are based on the RRBS platform. The output format is given Figure 2. 2) Analysis and annotation results for the methylated CpG sites (mCpG), which provide the methylation percentage value for all mCpG sites across the Bangladesh J Pharmacol 2016; 11: S154-S160 S155 Figure 1: Schematic illustration for the DMAK structure. The first level contains ENCODE data preprocess (namely, cell curation and data format process); the second level includes integrative analysis on the ENCODE data, namely DNA methylation CpGs annotation, identification of differentially-methylated CpGs and regions; the third level covers result visualization and further- multi-scale interrogation of biological functions mentioned cell lines. The higher percentage value indicates the higher methylation status, and vice versa (Figure 3). 3) Analysis and annotation results for the significant differentially-methylated CpG sites (SDMC) with reference to one cell type, here we selected T-47D as the study case. The results are further filtered based on the lifted methylation difference threshold (at least 25% methylation difference for the paired groups). And the SDMC list contains 106,252 DMCs (Akalin et al., 2012a; Akalin et al., 2012b), together the related statistical p-value and adjusted q-value are also provided in Figure 4. 4) Statistical analysis and annotation results for the differentially-methylated regions (DMR) with reference to one cell type, for consistence we selected T-47D as the case. We identified 16,277 DMR candidates from all the DMCs, with the adjusted q- value ≤0.01, CpG base methylation difference cut off, 25, and DMR mean methylation difference cut off, 20. Within those candidates, 8,936 entries present hyper- methylated and 7,341 with hypo-methylated status. With the lifted thresholds, namely adjusted q-value ≤0.001, differentially-methylated CpG base count ≥5, we further detected 7,537 significant DMRs (Sig- DMRs), where 3,512 entries are significantly hypermethylated-DMRs (Sig-Hyper-DMRs), and 4,025 significantly hypomethylated-DMRs (Sig-Hypo- DMRs). The output format is shown in Figure 5. 5) Statistical analysis and annotation results for the significantly hypermethylated-DMRs (Sig-Hyper- DMRs) with reference to T-47D cell type as shown in the output format (Figure 6). 6) Statistical analysis and annotation results for the significantly hypomethylated-DMRs (Sig-Hypo- DMRs) with reference to T-47D cell type (Figure 7). Figure 2: Schematic illustration of statistical information detected from RRBS sequencing read coverage, number of Cs and Ts Figure 3: Schematic illustration of methylated CpG sites across all listed cell types. The methylation value is of percentage format, annotated with purple bars for each cell line Figure 4: Schematic illustration of significantly DMC (SDMC). The methylation difference is annotated with blue bars for each CpG site in T-47D cell line Figure 5: Schematic illustration for the identified DMRs with reference to T-47D cell type. The table provides the width, mean methylation difference, the corresponding p-value and adjusted q-value for each DMR entry. The mean methylation difference is annotated with green or red bars for each DMR S156 Bangladesh J Pharmacol 2016; 11: S154-S160 7) Statistical analysis and annotation results for the identified genes from all DMRs (hyper-DMRs and hypo-DMRs) with reference to T-47D cell type (Figure 8). Visualization and Function Analysis for the Annotation Results This section discusses the visualization and function analysis for the annotated results. Together we seek to detect whether there exists any functional association (Szklarczyk et al., 2015) between those identified genes from hyper-DMRs and hypo-DMRs, which can explain the differential expression between those genes qualitatively, especially for the genes belonging to tumor suppressor genes (TSG) (Bedi et al., 2014; Blattler et al., 2014; Zhao et al., 2013). Thus we annotated the genes identified from DMRs with TSG information, filtered out those from unknown sources, and constructed the TSG functional association networks for hyper-DMR and hypo-DMR, respectively. For illustration and space limitation, Figure 8 depicts the 20-TSG functional association structures for hyper- and hypo-DMRs, respectively. For validating the high fidelity of the analysis results, those 20 TSGs are randomly selected from the TSG list for each case. And interestingly, we found most of those TSG nodes are functionally associated to form clusters. In hyper- DMR case, Figure 9A, only 4 out of 20 TSGs are dissociated from the TSG cluster; for hypo-DMR case, Figure 9B, it is comparatively loosely-connected and 10 out of 20 TSGs are not linked to the TSG cluster. The complete TSG functional association network structures for hyper-DMR and hypo-DMR are provided in DMAK package deposited at GitHub. The TSGs in those structures are highly physically connected and functional associated in DMRs for our T-47D breast cancer case. Figure 10 depicts the Gene Ontology (Sherman et al., 2007)analysis results for the two functional protein association network inferred for the TSGs. The upper (A) is for the hyper-DMRs, and the corresponding GO terms clearly prove such processes as transcription regulation, differentiation, mutation, activator, pathway in cancer and tumor suppressor, which are closed related to the hypermethylation outcomes of tumor suppressor genes. Figure 6: Schematic illustration for the identified hyper-DMR with reference to T-47D cell type. The mean methylation difference is annotated with red bars for each DMR Figure 7: Schematic illustration for the identified hypo-DMR with reference to T-47D cell type. The mean methylation difference is annotated with red bars for each DMR Figure 8: Schematic illustration for the identified gene information (SYMBOL and ENTREZ ID), log2 fold change, methylation percentage, tumor suppressor gene category (TRUE/FALSE), location (Promoter, CDS, Gene, 5’UTR, 3’UTR and Intron) and relat- ed methylation level (HYPER/HYPO) from DMRs with reference to T-47D cell type. The log2 fold change and methylation per- centage is annotated with red bars for each annotated gene Bangladesh J Pharmacol 2016; 11: S154-S160 S157 And the bottom (B) for the hypo-DMRs, and its GO terms present positive regulation of transcription and gene expression, differentiation, which to a certain extent confirm its connectivity to the hypomethylation outcomes of TSGs. Conclusion Our cross-cell DNA methylation annotation and analysis provide the systematic information knowledgebase for pan-cancer study. It contains curated reference results for ready-to-use information for sharing and rapid reanalysis. The first level of the knowledgebase is about raw data preprocess, we collected the data from the ENCODE Consortium portal. The second level is for annotation and function analysis; in this study case, we focused on DNA methylation in breast cancer cell, T-47D, annotated and identified the differentially-methylated sites and regions, and further identified the underlying tumor suppressor genes within the regions. The third level is for visualization and validation procedures. We further constructed the functional association network for the identified tumor suppressor genes, and further annotated the networks with Gene Ontology information, which can provide statistically significant evidences for the hyper-methylated and hypo- methylated processes in the breast cancer context. Our work provides a versatile and comprehensive platform for all biomedical researchers, especially for the genome-wide biomedical analysts, to interrogate and validate their hypothesis in an efficient and uniform way. In coming days, further annotation and analysis results concerning pan-cancer analysis will be updated into the knowledgebase, especially we seek to provide an interactive environment for biomedical researchers to fetch and utilize this knowledgebase. Acknowledgements The work was supported by the Research Fellowship from Harvard Medical School, the Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund (2nd Phase), the Fundamental Research Funds for China Central Universities under Grant Nos. 2012B03914 and XZX10B007-02, Changzhou Science & Technology Program under Grant No. CE20155050. This work made use of the Open Cloud Consortium (OCC)-sponsored project resource, which supported in part by grants from Gordon and Betty Moore Foundation and the National Science Foundation (USA) and major contributions from OCC members. Conflict of Interest The authors declare no conflict of interest. References Akalin A, Garrett-Bakelman FE, Kormaksson M, Busuttil J, Zhang L, Khrebtukova I, Milne TA, Huang Y, Biswas D, Hess JL, Allis CD, Roeder RG, Valk PJM, Löwenberg B, Delwel R, Fernandez HF, Paietta E, Tallman MS, Schroth GP, Mason CE, Melnick A, Figueroa ME. Base-pair resolution DNA methylation sequencing reveals profoundly divergent epigenetic landscapes in acute myeloid leukemia. PLoS Genet 2012a; 8: e1002781. Figure 9: Illustrative diagram for functional protein association network inferred for the tumor suppressor genes (TSG), identified from hyper-DMRs (A) and hypo-DMRs (B). The nodes represent the TSGs detected from DMRs, and links represent the associa- tion evidences. The full landscape for the TSG association networks in hyper-DMR and hypo-DMR are provided in the DMAK package deposited at GitHub S158 Bangladesh J Pharmacol 2016; 11: S154-S160 Akalin A, Kormaksson M, Li S, Garrett-Bakelman F, Figueroa M, Melnick A, Mason C. Methylkit. A comprehensive r package for the analysis of genome-wide DNA methylation profiles. Genome Biology 2012b; 13: R87. Bedi U, Mishra VK, Wasilewski D, Scheel C, Johnsen SA. Epigenetic plasticity: A central regulator of epithelial-to- mesenchymal transition in cancer. Oncotarget 2014; 5: 2016- 29. Blattler A, Yao L, Witt H, Guo Y, Nicolet C, Berman B, Farnham P. Global loss of DNA methylation uncovers intronic enhancers in genes showing expression changes. Genome Biology 2014; 15: 469. Bock C, Lengauer T: Computational epigenetics. Bioinforma- tics 2008; 24: 1-10. de Souza N. Genomics: The encode project. Nat Meth 2012; 9: 1046-1046. Kemp Christopher J, Moore James M, Moser R, Bernard B, Teater M, Smith Leslie E, Rabaia Natalia A, Gurley Kay E, Guinney J, Busch Stephanie E, Shaknovich R, Lobanenkov Victor V, Liggitt D, Shmulevich I, Melnick A, Filippova Galina N. Ctcf haploinsufficiency destabilizes DNA methylation and predisposes to cancer. Cell Reports 2014; 7: 1020-29. Kristensen VN, Lingjaerde OC, Russnes HG, Vollan HKM, Frigessi A, Borresen-Dale A-L. Principles and methods of integrative genomic analyses in cancer. Nat Rev Cancer 2014; 14: 299-313. Leiserson MDM, Vandin F, Wu H-T, Dobson JR, Eldridge JV, Thomas JL, Papoutsaki A, Kim Y, Niu B, McLellan M, Figure 10: The Gene Ontology (GO) analysis for the two functional protein association network inferred for the TSGs, the upper (A) for hyper-DMRs and the bottom (B) for the hypo-DMRs, sorted by the p-value Bangladesh J Pharmacol 2016; 11: S154-S160 S159 Author Info Binhua Tang (Principal contact) e-mail: bh.tang@outlook.com; Mobile: +86-18861231716 First two authors contributed equally Lawrence MS, Gonzalez-Perez A, Tamborero D, Cheng Y, Ryslik GA, Lopez-Bigas N, Getz G, Ding L, Raphael BJ. Pan- cancer network analysis identifies combinations of rare somatic mutations across pathways and protein complexes. Nat Genet 2015; 47: 106-14. Pennisi E. Encode project writes eulogy for junk DNA. Sci 2012; 337: 1159-61. Roadmap Epigenomics Consortium, Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, Heravi-Moussavi A, Kheradpour P, Zhang Z, Wang J, Ziller MJ, Amin V, Whitaker JW, Schultz MD, Ward LD, Sarkar A, Quon G, Sandstrom RS, Eaton ML, Wu Y-C, Pfenning AR, Wang X, Claussnitzer M, Liu Y, Coarfa C, Harris RA, Shoresh N, Epstein CB, Gjoneska E, Leung D, Xie W, Hawkins RD, Lister R, Hong C, Gascard P, Mungall AJ, Moore R, Chuah E, Tam A, Canfield TK, Hansen RS, Kaul R, Sabo PJ, Bansal MS, Carles A, Dixon JR, Farh K-H, Feizi S, Karlic R, Kim A-R, Kulkarni A, Li D, Lowdon R, Elliott G, Mercer TR, Neph SJ, Onuchic V, Polak P, Rajagopal N, Ray P, Sallari RC, Siebenthall KT, Sinnott- Armstrong NA, Stevens M, Thurman RE, Wu J, Zhang B, Zhou X, Beaudet AE, Boyer LA, De Jager PL, Farnham PJ, Fisher SJ, Haussler D, Jones SJM, Li W, Marra MA, McManus MT, Sunyaev S, Thomson JA, Tlsty TD, Tsai L-H, Wang W, Waterland RA, Zhang MQ, Chadwick LH, Bernstein BE, Costello JF, Ecker JR, Hirst M, Meissner A, Milosavljevic A, Ren B, Stamatoyannopoulos JA, Wang T, Kellis M. Roadmap Epigenomics C: Integrative analysis of 111 reference human epigenomes. Nature 2015; 518: 317-30. Sherman B, Huang D, Tan Q, Guo Y, Bour S, Liu D, Stephens R, Baseler M, Lane HC, Lempicki R. David knowledgebase: A gene-centered database integrating heterogeneous gene annotation resources to facilitate high-throughput gene functional analysis. BMC Bioinformatics. 2007; 8: 426. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, Simonovic M, Roth A, Santos A, Tsafou KP, Kuhn M, Bork P, Jensen LJ, von Mering C. String v10: Protein–protein interaction networks, integrated over the tree of life. Nucleic Acids Research 2015; 43: D447-52. Tang B, Wang X. Inferring genome-wide interplay landscape between DNA methylation and transcriptional regulation P J Pharm Sci 2015; 28: 349-52. The Cancer Genome Atlas Research Network, Weinstein JN, Collisson EA, Mills GB, Shaw KR, Ozenberger BA, Ellrott K, Shmulevich I, Sander C, Stuart JM. The cancer genome atlas pan-cancer analysis project. Nat Genet 2013; 45: 1113-20. The ENCODE Project Consortium: An integrated encyclopedia of DNA elements in the human genome. Nature 2012; 489: 57 -74. Witte T, Plass C, Gerhauser C. Pan-cancer patterns of DNA methylation. Genome Med. 2014; 6: 1-18. Zhao M, Sun J, Zhao Z: Tsgene. A web resource for tumor suppressor genes. Nucleic Acids Research 2013; 41: D970-76. Ziller MJ, Gu H, Muller F, Donaghey J, Tsai LTY, Kohlbacher O, De Jager PL, Rosen ED, Bennett DA, Bernstein BE, Gnirke A, Meissner A. Charting a dynamic DNA methylation landscape of the human genome. Nature 2013; 500: 477-81. S160 Bangladesh J Pharmacol 2016; 11: S154-S160 Your feedback about this paper 1. Number of times you have read this paper 2. Quality of paper 3. Your comments Excellent Good Moderate Not good R Baseler M Lane HC Lempicki R David knowledgebase: Author Info Binhua Tang Principal contact email bhtangoutlookcom Mobile 8618861231716 First two authors contributed equally: Text1: Group2: Off Dropdown3: [0] DatePrinted: This article was downloaded by you on: Mar 19, 2016