Comprehensive and Robust Tools for Analysis of Tumor Heterogeneity and Evolution

项目总结/摘要 近年来，基于个体患者的分子变化的精准医学方法 肿瘤已经成为诊断和治疗癌症有希望策略。这些方法 肿瘤是细胞的异质集合，随着时间的推移而变化， 对治疗的反应在DNA序列水平上，变化范围从单核苷酸 突变到大的染色体重排和全基因组复制。新的DNA/RNA 测序技术能够测量这种异质性，并提供数据来推断 肿瘤的进化史然而，分析所需的算法和软件 肿瘤异质性和演变的复杂性在范围上仍然有限。我们建议发展一个 全面的软件工具包，用于分析肿瘤异质性和肿瘤在空间，时间， 和基因组规模。这个工具包将基于先进的组合和统计算法 这是PI在过去几年中开发的。这些算法将统一成一个强大的， 计算效率高，统计可靠的软件包。该工具包将包含模块 对于不同类型的肿瘤样品，包括单个肿瘤样品、多个肿瘤区域、多个肿瘤 解剖部位（例如原发性肿瘤和转移）和多个时间点。软件还将 分析来自不同测序方法（全基因组、全外显子组和靶向测序）的数据 测序）和不同的测序技术，包括大块肿瘤、单细胞、短读段和 长读。该软件包将是开源的，并将在个人电脑上运行， 计算集群或云计算环境中。广泛的文件和培训将 以便于从生物信息学专家到临床医生的广泛用户使用。这些 强大的数据分析工具将使研究人员能够描述肿瘤内的异质性， 更高的准确性，使癌症诊断和治疗更加精确。

Project Summary/Abstract In recent years, precision medicine approaches based on molecular changes in an individual patient’s tumor have become a promising strategy for diagnosis and treatment of cancer. These approaches are challenged by the fact that tumors are a heterogeneous collection of cells that change over time and in response to treatment. At the DNA sequence level, changes range in scale from single-nucleotide mutations to large chromosomal rearrangements and whole-genome duplications. New DNA/RNA sequencing technologies enable measurement of this heterogeneity and provide data to infer the evolutionary history of a tumor. However, the algorithms and software necessary to analyze the complexities of tumor heterogeneity and evolution remain limited in scope. We propose to develop a comprehensive software toolkit to analyze tumor heterogeneity and tumor evolution across space, time, and genomic scale. This toolkit will be based on advanced combinatorial and statistical algorithms developed by PI over the past several years. These algorithms will be unified into a robust, computationally efficient, and statistically sound software package. This toolkit will incorporate modules for different types of tumor samples including single tumor samples, multiple tumor regions, multiple anatomical sites (e.g. primary tumor and metastasis), and multiple time points. The software will also analyze data from different sequencing approaches (whole-genome, whole-exome, and targeted sequencing) and different sequencing technologies including bulk tumor, single-cell, short-read, and long-read. The software package will be open source and will be released to run on individual computers, computing clusters, or in cloud computing environments. Extensive documentation and training will be provided to facilitate use by a wide range of users from expert bioinformaticians to clinicians. These powerful data analytic tools will enable researchers to characterize the heterogeneity within tumors with high accuracy, enabling greater precision in cancer diagnosis and treatment.