A Transformer based mutation prediction algorithm

سال انتشار: 1401
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 123

نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد

این مقاله در بخشهای موضوعی زیر دسته بندی شده است:

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

IBIS11_029

تاریخ نمایه سازی: 19 آذر 1402

چکیده مقاله:

Protein mutation prediction is the process of identifying the consequences of amino acid sequence alterations in a protein. It may be used to estimate the possible influence of genetic alterations on the function of a protein, as well as to discover disease-causing mutations or mutations that may impact therapeutic e cacy. Mutation prediction is often performed using machine learning models that analyze the functional impact of mutations based on reference data from evolutionary conservation, protein structure, and population genetics. Support Vector Machines (SVMs), Decision Trees, and Artificial Neural Networks are the most frequently used mutation prediction techniques. In addition, more advanced models, such as the Transformer model, have been used to predict mutations . In this work we developed a Transformer based mutation prediction algorithm to generate new mutants protein sequences based on a sequence that have already seen in nature which can be called a parent sequence. A transformer-based mutation prediction method is a machine learning model that predicts the impact of mutations on protein sequences using a transformer architecture. Transformers are predominantly employed for natural language processing tasks including language translation, question answering, as well as text summarization. The transformer architecture is based on an attention mechanism, which enables the model to concentrate on certain input components while generating predictions. This enables the model to better comprehend the input’s context, resulting in more precise predictions. From language translation to protein mutation prediction, transformers have been utilized for a variety of purposes. For the results, we discover out which sequence seems to be more probable to mutant in the future with the help of bioinformatic methods such as BLOSUM Score, and we may utilize these findings for preventing a potential pandemic.

نویسندگان

khashayar ehteshami

Sharif university of technology international campus kish island.

Seyed amirali Ghafourian ghahremani

Sharif university of technology international campus kish island

Kaveh Kavousi

University of tehran