Self-Interacting Proteins Prediction from PSSM Based on Evolutionary Information

Abstract

Self-interacting proteins (SIPs) play an influential role in regulating cell structure and function. Thus, it is critically important to identify whether proteins themselves interact with each other. Although there are some existing experimental methods for self-interaction recognition, the limitations of these methods are both expensive and time-consuming. Therefore, it is very necessary to develop an efficient and stable computational method for predicting SIPs. In this study, we develop an effective computational method for predicting SIPs based on rotation forest (RF) classifier, combined with histogram of oriented gradients (HOG) and synthetic minority oversampling technique (SMOTE). When performing SIPs prediction on yeast and human datasets, the proposed method achieves superior accuracies of 97.28% and 89.41%, respectively. In addition, the proposed approach was compared with the state-of-the-art support vector machine (SVM) classifiers and other different methods on the same datasets. The experimental results demonstrate that our method has good robustness and effectiveness and can be regarded as a useful tool for SIPs prediction.

SIPGCN: A Novel Deep Learning Model for Predicting Self-Interacting Proteins from Sequence Information Using Graph Convolutional Networks

“…In the model evaluation, these methods have achieved good results. Aiming at the SIP in PPI prediction problem, SMOTE [ 52 ], PSPEL [ 53 ], RP-FFT [ 44 ], SPAR [ 26 ], and LocFuse [ 54 ] have put forward better solutions to the problem. To better assess the capabilities of SIPGCN, we compared it with these models.…”

Large Span Sizes and Irregular Shapes Target Detection Methods Using Variable Convolution-Improved YOLOv8

“…And, most of the current technologies can only distinguish nonferrous metals or plastic sundries in scrap steel but fail to achieve the classification of scrap steel grades. Additionally, the classification of scrap steel into light as shown in Figure 1a, medium as shown in Figure 1b, heavy types as shown in Figure 1c, oily as shown in Figure 1d, and confined as shown in Figure 1e The traditional detection algorithm types are the V-J (Viola and Jones) detection algorithm [10], the HOG (Histogram of Oriented Gradients) and SVM (Support Vector Machine) detection algorithm [11], the DPM (Deformable Part-Based Model) algorithm [12], etc. However, there are still many problems with traditional target detection methods because sliding windows bring a large number of redundant windows, which consume a lot of time, and because the robustness and generalization are too poor due to the use of manually extracted features [13], so the traditional target detection algorithms are not suitable for use in industrialized scenarios.…”

“…Two primary categories of deep learningbased target detection algorithms exist: two-stage and one-stage. In contrast to the twostage approach, the one-stage algorithm employs a single network to directly predict The traditional detection algorithm types are the V-J (Viola and Jones) detection algorithm [10], the HOG (Histogram of Oriented Gradients) and SVM (Support Vector Machine) detection algorithm [11], the DPM (Deformable Part-Based Model) algorithm [12], etc. However, there are still many problems with traditional target detection methods because sliding windows bring a large number of redundant windows, which consume a lot of time, and because the robustness and generalization are too poor due to the use of manually extracted features [13], so the traditional target detection algorithms are not suitable for use in industrialized scenarios.…”