இயற்பியல் வேதியியல் & உயிர் இயற்பியல் இதழ்
திறந்த அணுகல்

சுருக்கம்

Specific Peptides Predict Protein Classification

David Horn*, Uri Weingart

The methodology of Specific Peptides (SP) has been introduced within the context of enzymes. It is based on unsupervised Machine Learning (ML) tool for motif extraction, followed by supervised annotation of motifs. In the case of enzymes, the classifier is the Enzyme Classification (EC) number. Here we restudy this problem, and demonstrate that we reach precision of 0.965 and recall of 0.891 on presently available protein sequences. Moreover, applying our methodology to query proteins is much faster than deep learning methods used for the same purpose.

We also apply this method to two other protein groups, G Protein Coupling Receptors (GPCR) and zinc finger proteins, find their corresponding SPs, and provide the code for searching any protein sequence for its classification under any such family. Some proteins which have annotations belonging to two of the three systems are being discussed. Our methodology can be applied to any protein group in order to find their corresponding SPs and provide the code for searching any protein sequence for its classification under any such family.