2020 Fiscal Year Research-status Report
Stem cell differentiation platform utilizing Bayesian machine learning
| Project/Area Number |
20KK0160
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
安井 孝介 京都大学, 高等研究院, 特定助教 (10877640)
ABDALKADER Rodi 京都大学, 高等研究院, 特定助教 (20839964)
|
|---|
| Project Period (FY) |
2020-10-27 – 2024-03-31
|
| Keywords | Stem cell / Cardiomyocite / Screening / Machine learning / DFT / Data-driven |
|---|
| Outline of Annual Research Achievements |
This project aims to develop a data-driven screening system for compounds which can stimulate stem cell differentiation into cardiomyocites. Due to difficulties with the on-going pandemic we are unable to travel overseas to acquire the necessary data at present. In order to overcome this problem, we therefore established connections with other stem cell researchers in Japan, and succeeded to acquire a small database of compounds with known activity in stem cells. Using this database we have started to build a machine learning method which can classify compounds based on their activity in stem cells. At present, our method incorporates kernelized classification methods with density functional theory techniques from physics (similar to approaches used in the materials informatics area).
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
During the first few months the project was delayed due to our inability to travel overseas as well as the unexpected loss of a stem cell laboratory at the host institute. However, after connecting with other stem researchers in Japan and receiving data from them, we could put the project back to on schedule.
|
| Strategy for Future Research Activity |
We will continue to develop our machine learning algorithm using the data acquired so far. We will also negotiate with our domestic partners in order to acquire new data. The PI hopes to travel to UCLA towards the end of FY2021 to acquire additional data, as per the original plan. We aim to have an adequate machine learning algorithm within the next 6 months.
In addition, we will expand our machine learning algorithm to include data from single-cell RNA sequencing experiments. These high-quality data will describe the maturity state of the cells following compound treatment in detail, and should improve the predictive abilities of the machine learning algorithm. If necessary, single-cell RNA sequencing data can be acquired from a core facility at UCLA.
|
| Causes of Carryover |
The items that we purchased were cheaper than expected. Next year, this budget will be added to the cost of reagents for experiment.
|
