CrossBind: Collaborative Cross-Modal Identification of Protein Nucleic-Acid-Binding Residues

Authors: Linglin Jing, Sheng Xu, Yifan Wang, Yuzhe Zhou, Tao Shen, Zhigang Ji, Hui Fang, Zhen Li, Siqi Sun

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experimental results demonstrate that our approach outperforms the next best state of the art methods, Graph Site and Graph Bind, on DNA and RNA datasets by 10.8/17.3% in terms of the harmonic mean of precision and recall (F1 Score) and 11.9/24.8% in Matthews correlation coefficient (MCC), respectively.
Researcher Affiliation Collaboration 1Shanghai Artifcial Intelligence Laboratory 2Department of Computer Science, Loughborough University 3Research Institute of Intelligent Complex Systems, Fudan University 4 SSE & FNII, The Chinese University of Hong Kong (Shenzhen) 5Shanghai Jiao Tong University
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes We release the code at https://github.com/BEAM-Labs/Cross Bind.
Open Datasets Yes We utilized two benchmark datasets, DNA 129 and RNA 117 dataset, from a previous study for training and testing our method. These datasets were obtained from the Bio Lip database (Yang, Roy, and Zhang 2012) and consist of experimentally determined complex structures.
Dataset Splits Yes We trained and validated our method on the training dataset, with a validation ratio of 0.1.
Hardware Specification No The paper does not specify any hardware details such as GPU models, CPU types, or memory used for running the experiments.
Software Dependencies No The paper mentions "Adam optimizer" and "cosine annealing" but does not specify version numbers for any software dependencies, libraries, or programming languages used (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes In all experiments, we used the Adam optimizer with a weight decay of 1 10 4 and employed cosine annealing as the learning rate scheduler. We set the initial learning rate to 1 10 3 for the ALS module and 1 10 4 for the cross-modal module. In the RPF module, we ranked the nearest neighbors and selected the top five amino acids for the propensity filter using prediction logits with thresholds of [-0.8, 0.8], corresponding to a positive over a negative propensity.