This is from my internship project. I used scanpy and developed configuration functions which saves time around 15%. To avoid bugs, I recommend you to read manual.
If you see this: "*", that part of program CAN be configured by that command.
If you see fully colored title "*!* Colored Title", that part of program MUST be configured.
You press "Run All" & program will stop according to "break_command" you defined. You configure the related path, change parameters & run again until you satisfy from the results.
break_command & move on.
- Define Path:
Reading Dataset(Break Point 1) - Check
Normalizing(Break Point 2) - Find PCA:
PCA & Drawing Graph Of The Potantial Clusters(Break Point 3) - Find Number Of Cluster:
Configuration Of Clustering(Break Point 4) - Check Ranked Genes:
Plotting Of Ranked Gene Groups (Default Graphs)(Break Point 5) - Annotate Part I:
Configuration & Plotting Part Of Annotation Part I(Break Point 6) - Annotate Part II:
Configuration & Plotting Part Of Annotation Part II(Break Point 7) - Inspect Reference Cell Type:
Find Resolution & Inspection Of Reference Cell Type Within Each Other(Break Point 8) - Inspect Target Cell Type:
Inspection Of Reference Cell Type With Target Cell Type(Break Point 9)
Program is finished when you see Finished You may leave your conclusions to there.
Must To Read: To run this program & anlayze, you need to have ranked_functions.py & gene_markers.py files in your dataset folder. If these are missig you can use sections after Finished section.
For more detailed changes, function and instructions of the related functions can be read.