Releases: zengjiexia/ACT
Release v2.2
Important update, please run setup_conda_env.bat to reinstall the environment.
Major updates
Dependency
- Python 3.8+ will be used for this and later version of ACT to accommodate the required packages.
- pyqtgraph is no longer included locally.
- Several packages had been upgraded or downgraded to avoid dependency clash.
Algorithm
- Skan package will be used as the primary length calculation package for clusters detected from super-resolution images. Both main branch length and total branch lenght will be reported. However, measuring length of the skelentonized cluster is not ideal for dot-like/round particles, it would be more accurate to use the major axis length in this case.
Fixed bugs and improvements
Diffraction limited image analysis
- Fixed error raised when loading result of a previous analysis.
Super resolution image analysis
- Removed redundant lines in cluster analysis. Fixed error when only 1 cluster present.
- Fixed image not closed after SR drift correction.
- Provided alternative method to check SR image frames
Liposome Assay analysis
- Fixed error caused by dependency updates.
- Included progress bar in prompt for samples.
- Changed result folder naming.
Miscellaneous
- Setup batch now will remove the existed ACT_python3 environment before build.
- Fixed UI connections.
- Muted FutureWarning from pandas.
- Updated simulation codes.
Release v2.1 - ACT Publication
Stable release of ACT for publication on Cell Reports Methods.
ACT: A Computational Suite for the Structural and Functional Characterisation of Amyloid Aggregates
Major bug fixed - SR - Cluster profiling
An error was reported for the cluster profile result calculated from the super-resolution analysis. The empty row and column in the reconstructed images with cluster id were droped. This led to a wrongly calculated cluster size, area, eccentricity, and hence length calculation. The error is fixed in this release.
Full Changelog: v2.0.1...v2.0.2
Release v2.0.1
- Improved error handling in SR analysis:
a. Skip the images which were damaged - failed to be loaded by imagej.io.
b. Fixed output type when result with no localisation was encountered. Avoided mission abort when no localisation or cluster detected in a field of view. - Fixed no response of UI when trying to cluster a dataset which was not super-resolved.
Release v2.0
Version 2 development completed
Features
- Particle detection from diffraction limited images
- ComDet method
- PyStar method
- Super-resolution image analysis
- Image reconstruction with GDSC SMLM 1 and ThunderStorm
- Drift correction with Cross-correlation and Fiducial correction
- DBSCAN for cluster analysis
- Particle profiling and length calculation
- Liposome assay analysis
- Tools
- Splitting images into multiple folders based on file names
Updates
- Updated python environment installation method
- User interface clean up
- Changed the program name to Aggregate Characterization Toolkit
Release v1.2.1
New feature:
Super-resolution image analysis
Debug:
- Diffraction-limit analysis edge particle removal parameter
- Diffraction-limit analysis method Trevor parameter
Release v1.1.1
- Enabled multiprocessing for Liposome assay.
- Minor bug correction.
Beta release for super-resolution image analysis
This is a pre-release containing function that may not work fully as expected.
This pre-release does not contain features updated in v1.1.0 onwards.
-
ThunderSTORM:
- Image reconstruction
- Fiducial correction with autocorrelation
- Fiducial correction with fiducial markers
-
GDSC SMLM 1:
- Image reconstruction
-
Bug correction:
- The wrong parameters were assigned for the diffraction-limited image analysis method Trevor.
Release v1.1.0
New features/improvements
Included liposome assay analysis in UKIA.
Upcoming features/improvements
- Improvement of orthogonal analysis.
- Single molecular image analysis.
Release v1.0.1
- ComDet - Remove edge particles detected.
- Trevor - Enabled multiprocessing.