%A Ghulam, Ali %A Sikander, Rahu %A Talpur, Dhani Bux %A Saba, Erum %A Talpur, Mir Sajjad Hussain %A Maher, Zulfikar Ahmed %A Tunio, Saima %D 2022 %T IDENTIFYING MOLECULAR FUNCTIONS OF DYNEIN MOTOR PROTEINS USING EXTREME GRADIENT BOOSTING ALGORITHM WITH MACHINE LEARNING %K %X The majority of cytoplasmic proteins and vesicles move actively primarily to dynein motor proteins, which are the cause of muscle contraction. Moreover, identifying how dynein are used in cells will rely on structural knowledge. Cytoskeletal motor proteins have different molecular roles and structures, and they belong to three superfamilies of dynamin, actin and myosin. Loss of function of specific molecular motor proteins can be attributed to a number of human diseases, such as Charcot-Charcot-Dystrophy and kidney disease.  It is crucial to create a precise model to identify dynein motor proteins in order to aid scientists in understanding their molecular role and designing therapeutic targets based on their influence on human disease. Therefore, we develop an accurate and efficient computational methodology is highly desired, especially when using cutting-edge machine learning methods. In this article, we proposed a machine learning-based superfamily of cytoskeletal motor protein locations prediction method called extreme gradient boosting (XGBoost). We get the initial feature set All by extraction the protein features from the sequence and evolutionary data of the amino acid residues named BLOUSM62. Through our successful eXtreme gradient boosting (XGBoost), accuracy score 0.8676%, Precision score 0.8768%, Sensitivity score 0.760%, Specificity score 0.9752% and MCC score 0.7536%.  Our method has demonstrated substantial improvements in the performance of many of the evaluation parameters compared to other state-of-the-art methods. This study offers an effective model for the classification of dynein proteins and lays a foundation for further research to improve the efficiency of protein functional classification. %U https://journal.kiu.edu.pk/index.php/JMAR/article/view/166 %J Journal of Mountain Area Research %0 Journal Article %R 10.53874/jmar.v8i0.166 %P 1-13%V 8 %@ 2518-850X %8 2022-11-29