METROID
METROID extracts and denoises transmembrane potential signals from single-cell fluorescence imaging to enable precise measurement of action potentials and electroporation events in low signal-to-noise conditions.
Key Features:
- Automated ROI Definition: Automatically defines regions of interest that adapt to cell morphology and follow cell contours to ensure consistent ROI size and shape for signal extraction from voltage-sensitive dyes and genetically encoded voltage sensors.
- Blind Source Separation (BSS) Techniques: Implements BSS methods to separate ROI signals from noise.
- PCA and ICA: Uses Principal Component Analysis (PCA) and Independent Component Analysis (ICA) as core decomposition methods for signal separation.
- Discrete Wavelet Transform (DWT) Variants: Provides two versions that incorporate discrete wavelet transform (DWT) to enhance separation of signal and noise.
- Single-Trial Signal Extraction: Filters subcellular fluorescence signals on a single-trial basis, removing the need for extensive signal averaging across repeated experiments.
- Low SNR Performance: Effectively extracts action potential (AP) and electroporation signals at signal-to-noise ratios as low as -10 dB without distorting the signal.
Scientific Applications:
- Membrane potential measurement: Enables precise measurement of membrane potentials in single-cell fluorescence experiments using voltage-sensitive dyes and genetically encoded voltage sensors.
- Fast subcellular event analysis: Supports analysis of rapid events such as action potentials and electroporation at subcellular resolution.
- Cell biology and biomedical research: Applies to basic cellular physiology studies and advanced biomedical research requiring accurate single-cell electrophysiological signals with minimal prior information.
Methodology:
Automated morphology-adaptive ROI definition; blind source separation using PCA and ICA; two DWT-incorporating variants; single-trial filtering of fluorescence images to extract AP and electroporation signals down to −10 dB SNR without signal distortion.
Topics
Details
- License:
- GPL-3.0
- Tool Type:
- command-line tool, desktop application
- Added:
- 1/18/2021
- Last Updated:
- 2/22/2021
Operations
Publications
Zoccoler M, de Oliveira PX. METROID: an automated method for robust quantification of subcellular fluorescence events at low SNR. BMC Bioinformatics. 2020;21(1). doi:10.1186/s12859-020-03661-9. PMID:32709217. PMCID:PMC7379836.
PMID: 32709217
PMCID: PMC7379836
Funding: - Fundação de Amparo à Pesquisa do Estado de São Paulo: Proc. N 2011/51199-6
Links
Repository
https://github.com/zoccoler/metroid