Development of Algorithm for Computational NMR-guided Magnetic Tweezer for Classification of Intermediates in DNA Interaction

Abstract

Deoxyribonucleic acid (DNA) is the molecule that carries genetic information of living organisms. DNA usually undergoes different interactions for molecular and cellular mechanisms, DNA modifications, transcription and recombination among others. One method to study the interactions DNA undergoes is to apply and measure the applied force on DNA using magnetic tweezer (MT). During DNA interaction, several intermediates occur and understanding these intermediates is crucial to properly classify DNA. Models to classify DNA intermediates are not readily available. In addition, it is sometimes difficult to measure large magnitude of force applied on biomolecules (e.g. DNA) using magnetic tweezer (MT) and this large amount of force could result in damage of DNA. Therefore, an algorithm that computationally classifies intermediates in DNA interaction is developed. The algorithm was developed using Python machine learning programming in relation to nuclear magnetic resonance (NMR) parameters similar to those obtainable in magnetic tweezers. With the developed algorithm in this study and machine learning commands, classification of intermediates in DNA interaction can be achieved by inputting the values of T₁ and T₂ relaxation times. The innovative model of classifying intermediates in DNA interaction is easy to use provided relaxation data (T₁ and T₂) is available. These data are obtained from relaxometers, which are easier to maintain compared to magnetic tweezers. In addition, computational means of classifying DNA intermediates helps to solve the problem of DNA damage as a result of large amounts of force applied. Hence, safety is ensured.