Treffer: DARUMA: a gateway to fast and easy prediction of intrinsically disordered regions.

Background: Intrinsically disordered proteins (IDPs) are proteins that contain intrinsically disordered regions (IDRs), which lack stable three-dimensional structures under physiological conditions. These regions are known to play crucial roles in many biological processes. While IDRs can be predicted from their amino acid sequences, and several accurate IDR prediction programs have been developed, such programs often require substantial computational resources, including long execution times, large databases for homology searches, and advanced computer architectures. Since DNA sequence data continues to grow rapidly, particularly at a genomic scale, there is an increasing need for fast and accurate IDR prediction programs that demand fewer computational resources. Methods: In this study, we developed DARUMA (Disorder order clAssifier by Rapid and User-friendly MAchine), an IDR prediction program designed for speed and ease of use. DARUMA uses a one-dimensional convolutional neural network (1D-CNN) that processes the physicochemical properties of amino acid residues instead of relying on sequence profiles. DARUMA employs a simple neural network that predicts IDRs using the output of 1D-CNN as input features. To ensure easy installation on users' systems, DARUMA was written entirely in Python using standard and NumPy libraries. Results: DARUMA achieves fast performance by avoiding iterative homology searches while delivering accuracy comparable to the latest predictors that use sequence profiles. In addition to the advantage of execution time, DARUMA requires no additional homology search programs and operates using standard Python libraries, making it easy to install and run on users' own environments without the need for specialized computational resources. DARUMA is available at https://antepontem.org/daruma/ , which also provides the stand-alone distribution. [ABSTRACT FROM AUTHOR]

Copyright of PeerJ Computer Science is the property of PeerJ Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)