PERCEPTRON – A Biology Breakthrough!
We are living, breathing chemical conglomerates who’ve carefully gathered knowledge over the course of centuries, to piece together clues in solving one of the greatest mysteries of our existence – the origin of life on Earth. An understanding of proteins has been a low hanging fruit of this endeavour. Decades of meticulous research on these magical molecules have revealed that much of life’s rich tapestry rests upon their complex and delicate structures. Therefore, tools and mechanisms that can zoom in on the structure of proteins and reveal the identification, are of incredible importance to scientists working on the frontier of life sciences research.
Dr. Safee Ullah Chaudhary, and his team of researchers, have developed a breakthrough tool, called PERCEPTRON, for highly efficient proteoform identification from top-down proetomics (TDP) spectral data. Their work has been published in the journal Nucleic Acids Research (NAR), a paramount achievement requiring nothing less than an outstanding team of researchers.
Reflecting on this distinction, Dr. Safee Ullah said "SBASSE at LUMS has brought together the right ingredients to undertake and sustain world-class research and development work. PERCEPTRON is one sure step to propel LUMS to the cutting edge of Proteomics and Bioinformatics research.”
Dr. Safee Ullah
PERCEPTRON, a web-based platform that outperforms all other similar tools by up to 135% in terms of reported proteins numbers and is ten times more efficient in terms of runtime.
In a technological masterstroke, Dr. Safee Ullah and his team have designed PERCEPTRON’s search pipeline to bring together algorithms for six different parameters:
- Intact protein mass tuning
- De novo sequence tags-based filtering
- Characterization of terminal as well as post-translational modifications
- Identification of truncated proteoforms,
- In silico spectral comparison
- Weight-based candidate protein scoring
To meet PERCEPTRON’s heavy demand of powerful and robust technology, GPU units by NVIDIA were used, along with Microsoft ASP.NET and ANGULAR Frameworks.
Highlighting the importance of team work in this effort, Ms. Kanzal Iman, a PhD student, who is part of Dr. Safee Ullah’s team, said “Working on PERCEPTRON has instilled in me that when you believe in consistent hard work, great things happen. Team PERCEPTRON is a great example of learning from each other and growing together.”
Another fellow co-first author of the paper, Ms Amna Ghafoor commented on the overall journey leading to the publication “The journey from a single “Hello World” to the entire GPU programming of PERCEPTRON has been a very bumpy ride. But thanks to the amazing team and blessing of Allah Almighty, we were able to pull it off Alhamdullilah.”
Commenting on the future support for this research, Dr. Safee Ullah said “We are looking forward to further support from the university to establish PERCEPTRON as a go-to platform for Proteome research across the globe."
PERCEPTRON aims to fill the voids in conventional protein identification software for TDP data, and make the process easier, more efficient and reliable. We congratulate Dr. Safee Ullah Chaudhary and his team for championing this multi-year effort through in-house resources at SBASSE, and pioneering Computational Biology and Bioinformatics research at the School.
Muhammad Farhan Khalid
Reference: Muhammad Farhan Khalid, Kanzal Iman, Amna Ghafoor, Mujtaba Saboor, Ahsan Ali, Urwa Muaz, Abdul Rehman Basharat, Taha Tahir, Muhammad Abubakar, Momina Amer Akhter, Waqar Nabi, Wim Vanderbauwhede, Fayyaz Ahmad, Bilal Wajid, Safee Ullah Chaudhary, PERCEPTRON: an open-source GPU-accelerated proteoform identification pipeline for top-down proteomics, Nucleic Acids Research, 2021, gkab368, https://doi.org/10.1093/nar/gkab368
More information on PERCEPTRON: https://perceptron.lums.edu.pk/index.html#/home