Logo Logo
Help
Contact
Switch language to German
Inferring the clonal identity of single cells from RNA-seq data with Unique Molecular Identifiers
Inferring the clonal identity of single cells from RNA-seq data with Unique Molecular Identifiers
Cancer is an evolutionary disease, in which heterogeneous populations of tumor cells can emerge, proliferate, and disappear depending on selective and neutral processes. This principle has been observed in many studies of acute myeloid leukemia (AML), which is the most common blood cancer in adults. Clonal heterogeneity and evolution have been proposed to play a role in the high relapse rate of this type of cancer. In order to understand this feature, it is crucial to have adequate clinical and experimental models that can provide enough data to elucidate the evolutionary history of a tumor, such as patient-derived xenografts (PDX). These models can be combined with high-resolution sequencing technologies, such as single-cell RNA-seq, to provide a detailed view of the heterogeneity and molecular features of the tumor. However, adequate analytical tools have to be applied and developed in order to fully exploit such datasets. Here I present the analysis of the clonal heterogeneity of an AML patient and the corresponding PDX model, which was treated with multiple rounds of chemotherapy. This model allowed to study the response of the tumor populations to the pressure induced by the therapy, and the possible evolutionary forces behind it. Datasets for these AML samples were generated with multiple types of sequencing methods, one of which was single-cell RNA sequencing. To enable the analysis of somatic mutations and clonal populations in this kind of data, I developed a software package, which is capable of extracting and proofreading variant sequences by making use of Unique Molecular Identifiers (UMIs), which are sequence barcodes that allow to distinguish reads that come from PCR amplification duplicates. The benefits of employing this proofreading approach for variant calling and for inferring the clonal identity of single cells were demonstrated. Finally, I applied to the analysis of the single-cell data of the AML PDX samples that were treated with chemotherapy, as well as other datasets with UMI-based sequencing.
Acute myeloid leukemia, cancer genomics, cancer evolution, intratumour heterogeneity, single-cell RNA-seq, bioinformatics, whole genome sequencing, whole exome sequencing, clonal heterogeneity
Valtierra Gutierrez, Ilse Ariadna
2020
English
Universitätsbibliothek der Ludwig-Maximilians-Universität München
Valtierra Gutierrez, Ilse Ariadna (2020): Inferring the clonal identity of single cells from RNA-seq data with Unique Molecular Identifiers. Dissertation, LMU München: Faculty of Biology
[img]
Preview
PDF
Valtierra_Gutierrez_Ilse_A.pdf

3MB

Abstract

Cancer is an evolutionary disease, in which heterogeneous populations of tumor cells can emerge, proliferate, and disappear depending on selective and neutral processes. This principle has been observed in many studies of acute myeloid leukemia (AML), which is the most common blood cancer in adults. Clonal heterogeneity and evolution have been proposed to play a role in the high relapse rate of this type of cancer. In order to understand this feature, it is crucial to have adequate clinical and experimental models that can provide enough data to elucidate the evolutionary history of a tumor, such as patient-derived xenografts (PDX). These models can be combined with high-resolution sequencing technologies, such as single-cell RNA-seq, to provide a detailed view of the heterogeneity and molecular features of the tumor. However, adequate analytical tools have to be applied and developed in order to fully exploit such datasets. Here I present the analysis of the clonal heterogeneity of an AML patient and the corresponding PDX model, which was treated with multiple rounds of chemotherapy. This model allowed to study the response of the tumor populations to the pressure induced by the therapy, and the possible evolutionary forces behind it. Datasets for these AML samples were generated with multiple types of sequencing methods, one of which was single-cell RNA sequencing. To enable the analysis of somatic mutations and clonal populations in this kind of data, I developed a software package, which is capable of extracting and proofreading variant sequences by making use of Unique Molecular Identifiers (UMIs), which are sequence barcodes that allow to distinguish reads that come from PCR amplification duplicates. The benefits of employing this proofreading approach for variant calling and for inferring the clonal identity of single cells were demonstrated. Finally, I applied to the analysis of the single-cell data of the AML PDX samples that were treated with chemotherapy, as well as other datasets with UMI-based sequencing.