The shiny-based software for Fragment Analysis(CE), especially for MSI analysis
The Windows version for directly used:
click here to download ACMSI
(Linux or Mac can run app.R with MSIanalysis.py to build a server)
| Software | Size Calling (322) | MSI Classification | |||
|---|---|---|---|---|---|
| MSI markers (725) | Samples (126) | ||||
| sensitivity | specificity | sensitivity | specificity | ||
| ACMSI | 99.07% | 96.30% | 99.64% | 100% | 100% |
| GeneMarker | 97.52% | - | - | - | - |
| PeakScanner | 97.52% (0.93%) | - | - | - | - |
| GeneMapper | 95.96% | - | - | - | - |
| coffalyser.NET | 46.27% | - | - | - | - |
The overall workflow shows:
The result of MSI classification:
The red portion represents peaks shared between tumor and normal tissues
The purple portion represents elevated peaks in tumor tissue compared to normal tissue
The black portion represents new peaks in tumor tissue
Followings show the UI of 3 main modules of ACMSI:
1. Plot of Fragment Analysis
2. Calibration of Size Calling
3. Automated Classification of MSI
