Develop Whole Image Quality Control and Rerun Plate 5 #33
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
There was a problem hiding this comment.
It LGTM @jenna-tomkinson, good job! I left some comments. Maybe I missed this, but did you include the QC data? I didn't see it in the data folder. I'll also look more into the QC data during model training.
1.cellprofiler_ic/image_quality_control/qc_figures/Plate_3_channels_blur_density.png
Outdated
Show resolved
Hide resolved
1.cellprofiler_ic/image_quality_control/qc_figures/Plate_5_per_site_flagged_fov.png
Outdated
Show resolved
Hide resolved
1.cellprofiler_ic/image_quality_control/qc_figures/Plate_5_platemap_flagged_fov_per_well.png
Outdated
Show resolved
Hide resolved
1.cellprofiler_ic/image_quality_control/scripts/1.evaluate_qc.py
Outdated
Show resolved
Hide resolved
1.cellprofiler_ic/image_quality_control/scripts/1.evaluate_qc.py
Outdated
Show resolved
Hide resolved
The QC output used for developing thresholds is included in the PR. As for QC-specific data, there is no specific data as we are removing images from being processed during IC, so the SQLite and parquet files downstream are updated to only include cells from the good-quality images. |
|
@MattsonCam Thank you lots for the review! I will be merging now! |
Develop Whole Image Quality Control and Rerun Plate 5
In this PR, I add whole image quality control to filter out poor quality images and rerun only Plate 5 to generate data to use for analysis and machine learning.