Analyzing Gene Expression Changes Over Time
Author Information
Author(s): Di Camillo Barbara, Toffolo Gianna, Nair Sreekumaran K, Greenlund Laura J, Cobelli Claudio
Primary Institution: University of Padova
Hypothesis
Can we effectively select differentially expressed genes in time series experiments with limited data?
Conclusion
Method 2 outperforms the other methods for short time series, while Method 3 is better for long time series.
Supporting Evidence
- Method 2 outperformed Method 1 in terms of precision and recall for short time series.
- Method 3 performed better than Method 2 for long time series.
- The study utilized synthetic data to evaluate the performance of the methods.
Takeaway
This study looks at how to find genes that change their activity over time, even when we don't have a lot of data. It shows that some methods work better than others depending on how much data we have.
Methodology
The study compares three methods for selecting differentially expressed genes using synthetic data and applies them to a real case study.
Potential Biases
Potential bias due to reliance on synthetic data and assumptions about error distributions.
Limitations
The methods may not perform well if the number of replicates is insufficient or if the error characteristics change over time.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website