HeapMS: An Automatic Peak-Picking Pipeline for Targeted Proteomic Data Powered by 2D Heatmap Transformation and Convolutional Neural Networks
Peak picking and quality assessment of multiple reaction monitoring (MRM) data require a considerable amount of human effort, particularly for low-abundance and high-interference signals. Although several peak-picking software packages are available, they frequently fail to detect low-quality peaks and do not report low-confidence cases. In addition, to identify uncertain or error cases, manual evaluations involving the visual examination of all chromatograms are still required. In this study, we presented HeapMS, a web service for quality assessment and artificial-intelligence-assisted peak picking of MRM chromatograms.
|
Computing nodes | Queuing jobs | Running jobs | Finished jobs | Select |
---|---|---|---|---|
Node 2(without GPU) | 0 | 0 | 9 | |
Node 3* | 1 | 0 | 64 | Use Node 3(For developing and inhouse use) |
Node 4**(without GPU) | 0 | 0 | 6 |
*Node 2 and Node 3 use our new models trained with 4 samples(OSCC-1, OSCC-2, OSCC-3 and CKD).
**Node 4 uses our new models trained with 4 samples(OSCC-1, OSCC-2, OSCC-3 and CKD) and without GPU.