School of Engineering Department of Chemical and Biological Engineering 68 Improving Data Analysis Methods for Shotgun Proteomics Supervisor: LAM Hei Ning / CBE Student: KHAN Asif / COMP Course: UROP 1100, Fall This report focuses on the improvement of data analysis methods in shotgun proteomics, specifically addressing the challenge of chimeric spectra. Chimeric spectra, resulting from the co-isolation and fragmentation of multiple peptides, pose a significant obstacle in accurate peptide identification. The report reviews existing methodologies, including the CharmeRT's double-search strategy and the AI-driven CHIMERYS algorithm. An implementation of the double-search strategy is also tested and evaluated, highlighting its strengths and weaknesses, particularly in peak removal and database-dependent peptide identification. The report also proposes future directions of leveraging deep learning in the task of chimeric spectrum detection, aiming to significantly enhance the accuracy and efficiency of peptide identification in shotgun proteomics. Improving Data Analysis Methods for Shotgun Proteomics Supervisor: LAM Hei Ning / CBE Student: CHOW Yuen Sum / BIEN Course: UROP 1000, Summer Proteomics involves the extensive examination of proteins, focusing on their structures and functions. Detecting peptide isotopic clusters, or "features", in MS1 spectra and matching them to MS/MS-based peptide identifications is crucial in data-dependent acquisition (DDA) proteomics. Various tools, including AlphaPept, IQMMA, and StPeter, have emerged, each with unique algorithms. This study compared these tools using the PXD006447 dataset from ProteomeXchange, evaluating them based on the coefficient of variation (CV) of feature intensities—a measure of consistency and reproducibility. AlphaPept and IQMMA, leveraging multiple detection algorithms, showed lower CV values, indicating higher consistency than StPeter's MS2-based features. These results highlight the importance of selecting appropriate quantification methods based on specific experimental contexts. Determination of Nutritional Value in Black Soldier Fly as a Potential Alternative Food Source for Pets Supervisor: LAM Leung Yuk Frank / CBE Co-supervisor: HU Xijun / CBE Student: CHANDRA Jones Edbert / BIOT Course: UROP 1000, Summer The world’s demand for protein resources for animal feeds and human consumption is predicted to increase significantly in the upcoming years which will lead to a protein source shortage. The study will investigate an alternative protein from an insect species, the black soldier fly for pet food. The study aims to provide a small-scale estimate of the costs and protocol use of black soldier fly larvae in the industry. The culturing process follows industry-based methods and black soldier fly larvae are subjected to drying methods before protein extraction. The protein extraction process helps approximate the market value of black soldier fly larvae protein content. Free-radical scavenging activity will be implemented to check mainly for 2-diphenyl1-picrylhydrazyl (DPPH) and other radicals.
RkJQdWJsaXNoZXIy NDk5Njg=