PhD Seminar • Bioinformatics • Quality Assessment of PointIso Generated Peptide Features and Model Extension for 4D Feature Detection

Please note: This PhD seminar will be given online.

Fatema Tuz Zohora, PhD candidate
David R. Cheriton School of Computer Science

Supervisor: Professor Ming Li

We perform a quality assessment of peptide features generated by PointIso, our novel deep learning model so that it becomes more reliable for biomarker discovery. Next, we evaluate its potential in Label Free Quantification (LFQ) pipeline. Finally, we extend our model for detecting four-dimensional peptide features (involves additional ‘ion-mobility’ dimension) and achieve 4% higher detection than existing algorithms on the human proteome dataset. This extension shows a way to make PointIso more generic so that it is applicable in a wide range of works in both proteomics and general object detection problems.