肿瘤的发生发展过程中遗传变异与染色质空间结构的关联研究

Many molecular mechanisms of tumor development and progression have been revealed by investigating the genome genetic and epigenetic alterations, including gene mutation, copy number, DNA methylation, histone modification, etc. However, the relationship between the high-order chromatin conformation and these alterations is yet to be fully addressed. Recent years, with the emerging chromatin conformation capture technology, studying the relationship between chromatin alteration and chromatin 3D structure becomes feasible. We analyzed plenty of tumor mutation data along with chromatin conformation data and found that co-mutated genes are tend to be co-localized in chromatin 3D space; we further verified this discovery via two MHCC liver cancer cell lines. This discovery leads that the chromatin conformation is the structural foundation of multiple mutations in tumor. Based on this, we plan to 1) fully investigate the spatial proximity between the chromatin points where tumor genetic and epigenetic alterations happen as well as “spatial co-mutation hotspot”; 2) the 3D locations of the alteration points in the nucleus; 3) the dynamic shift of the alteration positions during tumor progression. This research aim to reveal the relationship between chromatin 3D structure and tumor genetic and epigenetic alterations; it not only help in the understanding of tumor development and progression mechanism, but also help the theoretical study of tumor diagnosis, subtyping, prognosis, and therapy.

人们通过研究基因突变、拷贝数变化、DNA甲基化、组蛋白修饰等基因组遗传与表观遗传变化，揭示了许多肿瘤发生发展的分子机理，然而染色质高级空间构象与这些变化的关系甚少探索。近年来随着染色质构象俘获技术的发展为研究染色质高级空间构象与肿瘤遗传变异提供了条件。本课题组前期整合肿瘤基因突变谱与染色质空间结构数据对比分析，结果显示，多数肿瘤的共突变基因在染色质空间距离非常接近，并且在肝癌细胞株测序数据进一步验证，提示染色质空间距离是肿瘤多基因同时突变的结构基础。在此基础上，本课题拟深入研究1)肿瘤的变异位点之间的空间距离关联性及空间“变异热点”的寻找；2)变异位点与其在细胞核内的位置关联性；3)染色质空间结构与肿瘤发展过程中遗传变异演变的关联性。本研究所揭示的染色质三维结构与肿瘤遗传与表观遗传变异的关系不仅能更深入理解肿瘤发生发展机制，而且为肿瘤诊断、分型、预后和治疗奠定一定理论基础。

人们通过基因组学与表观遗传学揭示了许多肿瘤发生发展的分子机理，然而染色质高级空间构象与这些变化的关系甚少探索。近年来随着染色质构象俘获技术的发展为研究染色质高级空间构象与肿瘤遗传变异提供了条件。本次课题整合了肿瘤基因突变谱与染色质空间结构数据对比分析，结果显示，多数肿瘤的共突变基因在染色质空间距离非常接近，并且在肝癌细胞株测序数据进一步验证，提示染色质空间距离是肿瘤多基因同时突变的结构基础。在此基础上，我们深入研究了1)肿瘤的变异位点之间的空间距离关联性，发表了肿瘤在染色质3D空间上的“变异热点”概念；2)利用染色质3D结构与深度学习对肿瘤经行分型预测，发表了两个算法DeepGene与DeepCNA。3）发现了肿瘤新抗原在染色质3D结构上的规律性。本次课题达到和超过了预期目标，为肿瘤诊断、分型、预后和治疗奠定一定理论基础。