Extrapolating Pseudo-Time Series for SST Interneuron Differentiation by Comparing Single-Cell and qPCR Data
(2024) FYTM06 20232Department of Physics
- Abstract
- A lack of somatostatin expressing interneurons (SST INs) is known to be a factor in several neurodegenerative and neurospsychiatric diseases, such as Alzheimer disease, Parkinson’s disease and schizophrenia. Being able to efficiently obtain SST INs in vitro could enable a greater understanding and better treatment of these diseases. The development of SST INs is not yet fully understood, but it is assumed that gene regulatory networks (GRN) govern the differentiation process from stem cells to SST INs. These GRN can be explored with models developed using experimental data. Such models require information about the temporal behavior of target genes, time series, which need to be obtained from experimental data. Sequencing data on... (More)
- A lack of somatostatin expressing interneurons (SST INs) is known to be a factor in several neurodegenerative and neurospsychiatric diseases, such as Alzheimer disease, Parkinson’s disease and schizophrenia. Being able to efficiently obtain SST INs in vitro could enable a greater understanding and better treatment of these diseases. The development of SST INs is not yet fully understood, but it is assumed that gene regulatory networks (GRN) govern the differentiation process from stem cells to SST INs. These GRN can be explored with models developed using experimental data. Such models require information about the temporal behavior of target genes, time series, which need to be obtained from experimental data. Sequencing data on single-cell resolution provides information regarding gene expression in great detail, but is time and cost intensive compared to the sequencing of bulk samples. In this work, a method is suggested to extract pseudo-time series from only one single-cell data set, using time development information from bulk sequencing. The method was tested on a data set for SST IN differentiation and yields results that are consistent with other experimental findings. (Less)
- Popular Abstract
- Neurons, i.e. nerve cells, are the building block of our nervous systems and are crucial for signaling information through our body. There exist several types of neurons, one of them is somatostatin expressing interneurons. These neurons are often reduced in patients with, for example, Alzheimer disease, Parkinson’s disease or schizophrenia. Thus, if it would be possible to generate those neurons in a lab, one might be able to treat these diseases by transplanting the missing neurons into a patient, or by initiating a new production of those neurons inside a patient.
Although a lot of information regarding the development of somatostatin expressing neurons is already gathered, it is still not fully understood and the development of the... (More) - Neurons, i.e. nerve cells, are the building block of our nervous systems and are crucial for signaling information through our body. There exist several types of neurons, one of them is somatostatin expressing interneurons. These neurons are often reduced in patients with, for example, Alzheimer disease, Parkinson’s disease or schizophrenia. Thus, if it would be possible to generate those neurons in a lab, one might be able to treat these diseases by transplanting the missing neurons into a patient, or by initiating a new production of those neurons inside a patient.
Although a lot of information regarding the development of somatostatin expressing neurons is already gathered, it is still not fully understood and the development of the neurons in a laboratory setting is often inefficient. Here, computational modelling of the developmental processes can help to find factors which make the desired processes more efficient. To perform modelling, it is required to obtain data on gene expression levels throughout the developmental processes. Gene expressions are a measurement of mRNA levels in cells. Hence, they can monitor in which state the cells are or what type of cells are present.
Currently, there are two different types of data that are commonly used to research cell development in a laboratory. One of them is bulk data. This data is relatively easy to obtain and consists of average gene expression values of many cells for a few pre-selected genes. The second type is single-cell data. For single-cell data, the gene expressions from individual cells are documented. Here, one data set typically contains individual information from several thousands of cells and expression values of 20000 to 30000 genes. Hence, single-cell data provides much more information about individual cells than bulk data, but is also more costly to obtain.
This thesis investigates a method that uses only one single-cell data set from the last day of the experiment and bulk data from different time points during the experiment. Therewith, it aims to minimize costs while still profiting from the detailed information from single-cell data. By combining the information from both data sets, it was possible to extract temporal information about the developmental process from the single-cell data, without the need for single-cell data from earlier time points.
After investigating this method, it can be concluded that it is possible to obtain information for a reasonable temporal development of gene expression from only one single-cell data set, with help of additional bulk data. Further, the method's potential to identify cells of interest in single-cell data in an unbiased manner is demonstrated.
More research is required to test the method on different data sets and to compare its performance to analyses using solely single-cell data or solely bulk data. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9169371
- author
- Hölldobler, Anna Lena LU
- supervisor
- organization
- course
- FYTM06 20232
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- pseudo-time series, single-cell data, single-cell analysis, qPCR data, time series analysis, SST Interneurons, cell differentiation
- language
- English
- id
- 9169371
- date added to LUP
- 2024-07-02 09:31:04
- date last changed
- 2024-07-05 13:34:07
@misc{9169371,
abstract = {{A lack of somatostatin expressing interneurons (SST INs) is known to be a factor in several neurodegenerative and neurospsychiatric diseases, such as Alzheimer disease, Parkinson’s disease and schizophrenia. Being able to efficiently obtain SST INs in vitro could enable a greater understanding and better treatment of these diseases. The development of SST INs is not yet fully understood, but it is assumed that gene regulatory networks (GRN) govern the differentiation process from stem cells to SST INs. These GRN can be explored with models developed using experimental data. Such models require information about the temporal behavior of target genes, time series, which need to be obtained from experimental data. Sequencing data on single-cell resolution provides information regarding gene expression in great detail, but is time and cost intensive compared to the sequencing of bulk samples. In this work, a method is suggested to extract pseudo-time series from only one single-cell data set, using time development information from bulk sequencing. The method was tested on a data set for SST IN differentiation and yields results that are consistent with other experimental findings.}},
author = {{Hölldobler, Anna Lena}},
language = {{eng}},
note = {{Student Paper}},
title = {{Extrapolating Pseudo-Time Series for SST Interneuron Differentiation by Comparing Single-Cell and qPCR Data}},
year = {{2024}},
}