Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GBDTSVM: Combined Support Vector Machine and Gradient Boosting Decision Tree Framework for efficient snoRNA-disease association prediction

Ummay Maria Muna, Fahim Hafiz, Shanta Biswas, Riasat Azim

TL;DR

The paper tackles the prediction of snoRNA-disease associations (SDA) amid limited labeled data and imbalanced classes by introducing GBDTSVM, a two-stage framework that first uses a Gradient Boosting Decision Tree to transform features and then applies a Support Vector Machine with an RBF kernel for classification. It enriches representations through integrated similarity measures, including snoRNA functional similarity, disease semantic similarity, and Gaussian Interaction Profile-based kernels, culminating in meshed similarity profiles that feed the GBDT. Empirical results on multiple datasets (MDRF, LSGT, PsnoD) show superior AUROC (≈0.96) and AUPRC (≈0.95) compared with seven SOTA methods, with additional validation via case studies against RNADisease and PubMed. The approach offers a scalable, interpretable alternative to deep learning models for SDA prediction and could extend to other ncRNA-disease prediction tasks, accelerating biomarker discovery and therapeutic target identification.

Abstract

Small nucleolar RNAs (snoRNAs) are increasingly recognized for their critical role in the pathogenesis and characterization of various human diseases. Consequently, the precise identification of snoRNA-disease associations (SDAs) is essential for the progression of diseases and the advancement of treatment strategies. However, conventional biological experimental approaches are costly, time-consuming, and resource-intensive; therefore, machine learning-based computational methods offer a promising solution to mitigate these limitations. This paper proposes a model called 'GBDTSVM', representing a novel and efficient machine learning approach for predicting snoRNA-disease associations by leveraging a Gradient Boosting Decision Tree (GBDT) and Support Vector Machine (SVM). 'GBDTSVM' effectively extracts integrated snoRNA-disease feature representations utilizing GBDT and SVM is subsequently utilized to classify and identify potential associations. Furthermore, the method enhances the accuracy of these predictions by incorporating Gaussian kernel profile similarity for both snoRNAs and diseases. Experimental evaluation of the GBDTSVM model demonstrated superior performance compared to state-of-the-art methods in the field, achieving an area under the receiver operating characteristic (AUROC) of 0.96 and an area under the precision-recall curve (AUPRC) of 0.95 on MDRF dataset. Moreover, our model shows superior performance on two more datasets named LSGT and PsnoD. Additionally, a case study on the predicted snoRNA-disease associations verified the top 10 predicted snoRNAs across nine prevalent diseases, further validating the efficacy of the GBDTSVM approach. These results underscore the model's potential as a robust tool for advancing snoRNA-related disease research. Source codes and datasets our proposed framework can be obtained from: https://github.com/mariamuna04/gbdtsvm

GBDTSVM: Combined Support Vector Machine and Gradient Boosting Decision Tree Framework for efficient snoRNA-disease association prediction

TL;DR

The paper tackles the prediction of snoRNA-disease associations (SDA) amid limited labeled data and imbalanced classes by introducing GBDTSVM, a two-stage framework that first uses a Gradient Boosting Decision Tree to transform features and then applies a Support Vector Machine with an RBF kernel for classification. It enriches representations through integrated similarity measures, including snoRNA functional similarity, disease semantic similarity, and Gaussian Interaction Profile-based kernels, culminating in meshed similarity profiles that feed the GBDT. Empirical results on multiple datasets (MDRF, LSGT, PsnoD) show superior AUROC (≈0.96) and AUPRC (≈0.95) compared with seven SOTA methods, with additional validation via case studies against RNADisease and PubMed. The approach offers a scalable, interpretable alternative to deep learning models for SDA prediction and could extend to other ncRNA-disease prediction tasks, accelerating biomarker discovery and therapeutic target identification.

Abstract

Small nucleolar RNAs (snoRNAs) are increasingly recognized for their critical role in the pathogenesis and characterization of various human diseases. Consequently, the precise identification of snoRNA-disease associations (SDAs) is essential for the progression of diseases and the advancement of treatment strategies. However, conventional biological experimental approaches are costly, time-consuming, and resource-intensive; therefore, machine learning-based computational methods offer a promising solution to mitigate these limitations. This paper proposes a model called 'GBDTSVM', representing a novel and efficient machine learning approach for predicting snoRNA-disease associations by leveraging a Gradient Boosting Decision Tree (GBDT) and Support Vector Machine (SVM). 'GBDTSVM' effectively extracts integrated snoRNA-disease feature representations utilizing GBDT and SVM is subsequently utilized to classify and identify potential associations. Furthermore, the method enhances the accuracy of these predictions by incorporating Gaussian kernel profile similarity for both snoRNAs and diseases. Experimental evaluation of the GBDTSVM model demonstrated superior performance compared to state-of-the-art methods in the field, achieving an area under the receiver operating characteristic (AUROC) of 0.96 and an area under the precision-recall curve (AUPRC) of 0.95 on MDRF dataset. Moreover, our model shows superior performance on two more datasets named LSGT and PsnoD. Additionally, a case study on the predicted snoRNA-disease associations verified the top 10 predicted snoRNAs across nine prevalent diseases, further validating the efficacy of the GBDTSVM approach. These results underscore the model's potential as a robust tool for advancing snoRNA-related disease research. Source codes and datasets our proposed framework can be obtained from: https://github.com/mariamuna04/gbdtsvm
Paper Structure (19 sections, 18 equations, 3 figures, 6 tables)

This paper contains 19 sections, 18 equations, 3 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Dataset
  4. Method
  5. Association Prediction
  6. Known snoRNA Disease Association
  7. snoRNA functional similarity
  8. Disease Semantic similarity
  9. Similarity enhancement of snoRNAs and diseases using Gaussian Kernel Profile similarity
  10. Gaussian Interaction Profile Kernel Disease Similarity
  11. Gaussian Interaction Profile Kernel snoRNA Similarity
  12. Meshed Similarity for snoRNAs and diseases
  13. Feature Extraction using Gradient Boosting Decision Tree
  14. Final Classification by Support Vector Machine
  15. Evaluation Measures
  16. ...and 4 more sections

Figures (3)

  • Figure 1: Initialy snoRNA features are utilized to derive snoRNA functional similarity by applying cosine similarity. Meanwhile, disease semantic similarities are measured from the disease DAG graph. Then gaussian kernel profile similarities were measured from the Association matrix for both of the similarity profiles and concatenated and divided into positive and negative samples that were the inputs of the GBDT model after being balanced by applying the clustering mechanism. Finally, the feature vectors generated from the GBDT model are trained using the SVM model and classified the snoRNA-disease association pairs.
  • Figure 2: The ROC and P-R Curves of GBDTSVM for snoRNA and disease association identification on MDRF dataset.
  • Figure 3: Performance Comparison with SOTA methods in 5-CV showing that GBDTSVM outperforms in terms of AUC and AUPR score across all the methods.