Transcription Factor-associated Combinatorial Epigenetic Pattern Of TCF7L2

With the coming of human post-genomic era,computer technology has been the most effective way on data mining and pattern recognition.Using the knowledge of molecular biology,we have already dissected biological problems from “hypotheses” to“data mining”,that is to integrate large quantities of biological data and discover the important biological mechanism.Numerous public data resources,such as the ENCODE,have generated thousands of genome-wide data sets and provided us with substantial quantities of data to study transcriptional and epigenetic patterns in different cell types at genome-wide scale with NGS techology.Transcription process in eukaryotes is highly regulated and is the major control step of gene expression.The transcriptional regulation is so complex,not only involves transcription factors and also be regulated by chromatin.Recently,studies have suggested that combinations of multiple epigenetic modifications are essential for controlling gene expression.Despite numerous computational approaches have been developed to decipher the combinatorial epigenetic patterns or “epigenetic code”,none of them has explicitly addressed the relationship between a specific transcription factor(TF)and the other epigenetic patterns.TCF7L2(transcription factor 7-like 2),an important component of WNT pathway,has been reported associated with several kinds of diseases including type 2 diabetes,carcinogenesis and bipolar disorder.The WNT pathway is often constitutively activated in human cancers,with high up-regulation of TCF7L2,especially in colon,breast and pancreatic cancer.Several studies have shown tissue-specific alternative splicing of TCF7L2,suggesting that TCF7L2 may have different functional properties in different cells.In our previous studies,we have mapped genome-wide binding of TCF7L2 in six cell lines.However,in these six cancer types the combinatorial epigenetic profile of TCF7L2 has not been well studied.In this study,TCF7L2-omics data was studied to identify combinatorial epigenetic patterns associated with cell type-specific transcription factors targets in different cell types.Unexpectedly,our method has uncovered a novel set of TCF7L2-regulatedintragenic enhancers in genome-wide.Furthermore,we validated the the accuracy and efficiency of prediction by molecular experimental technology.Throughout the entire study,the main works in this thesis are described as following:1.we developed a novel computational method,T-cep(Transcription factorassociated combinatorial epigenetic patterns),to identify TF-associated chromatin states and introduced the algorithm description and workflow.Comparing it with other published software tools,we found our tool has unique advantages underlying algorithmic design.On the one hand,it can ensure the signal purity of sample datasets.On the other hand,we could get more transcription factor associated combinatorial epigenetic patterns without decrease the accuracy of the result.2.Applying the next-generation sequencing to establish TCF7L2-omics data sets.We evaluated T-cep on the TCF7L2-omics data sets.Our results show that the method has uncovered three transcription factor associated combinatorial epigenetic patterns,including a novel set of TCF7L2-regulated intragenic enhancers missed by other software tools.Furthermore,we analyzed these three regulatory elements,including TCF7L2-associated promoter,TCF7L2-associated intragenic enhancer and TCF7L2-associated distal enhancer.We found the associated genes exert the highest gene expression of the TCF7L2 intragenic enhancer and cell type specific loci in five cancer cell lines.With GO analysis and KEGG pathway analysis,we predicted the TCF7L2 regulated intragenic enhancers may be more relevant to metastasis and disease progression with cancer type specificity.3.In order to validate the prediction,we further chose 17 genes of TCF7L2-associated intragenic enhancers and TCF7L2-associated distal enhancers in cell lines of MCF7 and PANC1.Applying si RNA knockdown and Co-transfection,we change the expression of TCF7L2,using RT-q PCR and Luciferase Reporter Assay technology to conduct and detect the experimental validation.The experimental results proved the accuracy and efficiency of prediction by T-cep as well as confirm the functionality of TCF7L2-regulated enhancers in both MCF7 and PANC1 cells respectively.4.For a broader application,we used MYC-omics data sets and identified five MYC-associated states,the similar pattern found in TCF7L2 data sets.Then we chose Chrom HMM,a powerful tools in genomic segmentation with a user-friendly application to compare with our tools.TCF7L2-omics datasets and MYC-omics datasets are trainedon Chrom HMM and we found Chrom HMM can identify more regulatory elements combinational pattern variants,however,missing the intragenic transcription factor enhancer state in both data sets,which found in the result of T-cep.Our method can find non-TF-combinational patterns as well as identified more transcription factor associated regulatory elements.