Protein Interaction Prediction Method Based on Feature Engineering and XGBoost
Human protein interaction prediction studies occupy an important place in systems biology. The understanding of human protein interaction networks and interactome will provide important insights into the regulation of developmental, physiological and pathological processes. In this study, we propose...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2023-01-01
|
| Series: | BIO Web of Conferences |
| Online Access: | https://www.bio-conferences.org/articles/bioconf/pdf/2023/06/bioconf_fbse2023_01021.pdf |
| _version_ | 1797776016161112064 |
|---|---|
| author | Zhao Xiaoman Wang Xue |
| author_facet | Zhao Xiaoman Wang Xue |
| author_sort | Zhao Xiaoman |
| collection | DOAJ |
| description | Human protein interaction prediction studies occupy an important place in systems biology. The understanding of human protein interaction networks and interactome will provide important insights into the regulation of developmental, physiological and pathological processes. In this study, we propose a method based on feature engineering and integrated learning algorithms to construct protein interaction prediction models. Principal Component Analysis (PCA) and Locally Linear Embedding (LLE) dimensionality reduction methods were used to extract sequence features from the 174-dimensional human protein sequence vector after Normalized Difference Sequence Feature (NDSF) encoding, respectively. The classification performance of three integrated learning methods (AdaBoost, Extratrees, XGBoost) applied to PCA and LLE features was compared, and the best combination of parameters was found using cross-validation and grid search methods. The results show that the classification accuracy is significantly higher when using the linear dimensionality reduction method PCA than the nonlinear dimensionality reduction method LLE. the classification with XGBoost achieves a model accuracy of 99.2%, which is the best performance among all models. This study suggests that NDSF combined with PCA and XGBoost may be an effective strategy for classifying different human protein interactions. |
| first_indexed | 2024-03-12T22:43:54Z |
| format | Article |
| id | doaj.art-1a28512564474480ab0de8168c66e4bb |
| institution | Directory Open Access Journal |
| issn | 2117-4458 |
| language | English |
| last_indexed | 2024-03-12T22:43:54Z |
| publishDate | 2023-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | BIO Web of Conferences |
| spelling | doaj.art-1a28512564474480ab0de8168c66e4bb2023-07-21T09:24:56ZengEDP SciencesBIO Web of Conferences2117-44582023-01-01610102110.1051/bioconf/20236101021bioconf_fbse2023_01021Protein Interaction Prediction Method Based on Feature Engineering and XGBoostZhao Xiaoman0Wang Xue1Institute of Intelligent Machinery, Hefei Institutes of Physical Science, Chinese Academy of SciencesInstitute of Intelligent Machinery, Hefei Institutes of Physical Science, Chinese Academy of SciencesHuman protein interaction prediction studies occupy an important place in systems biology. The understanding of human protein interaction networks and interactome will provide important insights into the regulation of developmental, physiological and pathological processes. In this study, we propose a method based on feature engineering and integrated learning algorithms to construct protein interaction prediction models. Principal Component Analysis (PCA) and Locally Linear Embedding (LLE) dimensionality reduction methods were used to extract sequence features from the 174-dimensional human protein sequence vector after Normalized Difference Sequence Feature (NDSF) encoding, respectively. The classification performance of three integrated learning methods (AdaBoost, Extratrees, XGBoost) applied to PCA and LLE features was compared, and the best combination of parameters was found using cross-validation and grid search methods. The results show that the classification accuracy is significantly higher when using the linear dimensionality reduction method PCA than the nonlinear dimensionality reduction method LLE. the classification with XGBoost achieves a model accuracy of 99.2%, which is the best performance among all models. This study suggests that NDSF combined with PCA and XGBoost may be an effective strategy for classifying different human protein interactions.https://www.bio-conferences.org/articles/bioconf/pdf/2023/06/bioconf_fbse2023_01021.pdf |
| spellingShingle | Zhao Xiaoman Wang Xue Protein Interaction Prediction Method Based on Feature Engineering and XGBoost BIO Web of Conferences |
| title | Protein Interaction Prediction Method Based on Feature Engineering and XGBoost |
| title_full | Protein Interaction Prediction Method Based on Feature Engineering and XGBoost |
| title_fullStr | Protein Interaction Prediction Method Based on Feature Engineering and XGBoost |
| title_full_unstemmed | Protein Interaction Prediction Method Based on Feature Engineering and XGBoost |
| title_short | Protein Interaction Prediction Method Based on Feature Engineering and XGBoost |
| title_sort | protein interaction prediction method based on feature engineering and xgboost |
| url | https://www.bio-conferences.org/articles/bioconf/pdf/2023/06/bioconf_fbse2023_01021.pdf |
| work_keys_str_mv | AT zhaoxiaoman proteininteractionpredictionmethodbasedonfeatureengineeringandxgboost AT wangxue proteininteractionpredictionmethodbasedonfeatureengineeringandxgboost |