Related Experiment Video
Updated: Aug 30, 2025
07:37
An Integrated Raman Spectroscopy and Mass Spectrometry Platform to Study Single-Cell Drug Uptake, Metabolism, and Effects
Published on: January 9, 2020
9.6K
Machine Learning-Based Determination of Sampling Depth for Complex Environmental Systems: Case Study with Single-Cell
Guangyu Li1,2, Chieh Wu3, Dongqi Wang1,4
1Department of Civil and Environmental Engineering, Northeastern University, Boston, Massachusetts 02115-5026, United States.
Environmental Science & Technology
|September 1, 2022
Summary
Determining sample size for microbial ecology studies is challenging. This research introduces a computational protocol using kernel divergence to optimize sample size, enhancing reproducibility in single-cell technology applications.
Area of Science:
- Microbial Ecology
- Computational Biology
- Analytical Chemistry
Background:
- Advanced analytical methods like single-cell technologies offer deeper microbial ecology insights.
- Determining adequate sample size is a major challenge due to unknown community complexity and resource limitations, hindering standardization.
Purpose of the Study:
- To propose, test, and validate a computational sampling size assessment protocol.
- To enable accurate determination of sample size for capturing desired information capacity and resolution levels.
Main Methods:
- Developed a computational protocol utilizing kernel divergence, a metric comparing dataset distributional differences.
- The method requires no human intervention or prior knowledge-based preclassification and makes minimal assumptions about data distribution.
- Validated the protocol using Single-cell Raman Spectroscopy (SCRS) data from enhanced biological phosphorus removal (EBPR) activated sludge communities.
Main Results:
- The kernel divergence metric effectively assesses distributional differences between datasets.
- The validated protocol allows for the determination of sufficient sample size for various information capture goals.
- Demonstrated flexibility in handling datasets with both linear and nonlinear relationships.
Conclusions:
- The proposed computational protocol offers a standardized approach for sampling size optimization in microbial ecology.
- Enhances comparability and reproducibility of experiments and analyses involving complex environmental samples.
- The method's flexibility allows extension to other single-cell technologies and environmental applications with continuous data features.
