Deep learning assisted fluid volume calculation for assessing anti-vascular endothelial growth factor effect in diabetic macular edema.

Objective: To develop an algorithm using deep learning methods to calculate the volume of intraretinal and subretinal fluid in optical coherence tomography (OCT) images for assessing diabetic macular edema (DME) patients' condition changes. Design: Cross-sectional study. Participants: Treatment-naive patients diagnosed with DME recruited from April 2020 to November 2021. Methods: The deep learning network, which was built for autonomous segmentation utilizing an encoder-decoder network based on the U-Net architecture, was used to calculate the volume of intraretinal fluid (IRF) and subretinal fluid (SRF). The alterations of retinal vessel density and thickness, and the correlation between best-corrected visual acuity (BCVA) and OCT parameters were analyzed. Results: 2,955 OCT images of fourteen eyes from DME patients with IRF and SRF who received anti-vascular endothelial growth factor (VEGF) agents were obtained. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve of the algorithm was 0.993 for IRF and 0.998 for SRF. The volumes of IRF and SRF were significantly decreased from 1.93 ± 0.58 /1.14 ± 0.25 mm3 (baseline) to 0.26 ± 0.13 /0.26 ± 0.18 mm3 (post-injection), respectively (p = 0.0170 for IRF, and p = 0.0004 for SRF). The Spearman correlation demonstrated that the reduction of IRF volume was negatively correlated with age (coefficient = -0.698, p = 0.006). Conclusion: We developed a deep learning assisted fluid volume calculation algorithm with high sensitivity and specificity for assessing the volume of IRF and SRF in DME patients. Key words: deep learning; diabetic macular edema; optical coherence tomography.

LPI-SKMSC: Predicting LncRNA-Protein Interactions with Segmented k-mer Frequencies and Multi-space Clustering.

 Long noncoding RNAs (lncRNAs) have significant regulatory roles in gene expression. Interactions with proteins are one of the ways lncRNAs play their roles. Since experiments to determine lncRNA-protein interactions (LPIs) are expensive and time-consuming, many computational methods for predicting LPIs have been proposed as alternatives. In the LPIs prediction problem, there commonly exists the imbalance in the distribution of positive and negative samples. However, there are few existing methods that give specific consideration to this problem. In this paper, we proposed a new clustering-based LPIs prediction method using segmented k-mer frequencies and multi-space clustering (LPI-SKMSC). It was dedicated to handling the imbalance of positive and negative samples. We constructed segmented k-mer frequencies to obtain global and local features of lncRNA and protein sequences. Then, the multi-space clustering was applied to LPI-SKMSC. The convolutional neural network (CNN)-based encoders were used to map different features of a sample to different spaces. It used multiple spaces to jointly constrain the classification of samples. Finally, the distances between the output features of the encoder and the cluster center in each space were calculated. The sum of distances in all spaces was compared with the cluster radius to predict the LPIs. We performed cross-validation on 3 public datasets and LPI-SKMSC showed the best performance compared to other existing methods. Experimental results showed that LPI-SKMSC could predict LPIs more effectively when faced with imbalanced positive and negative samples. In addition, we illustrated that our model was better at uncovering potential lncRNA-protein interaction pairs.

De novo miRNA-lncRNA Interactions Prediction via K-Square Encoding and Mixed Network.

Non-coding RNAs play a crucisal role in regulating various biological activities such as genetics and metabolism in plants. Traditional biological methods suffer from long research cycles and high costs. In recent years, bioinformatics methods combining deep learning have mainly focused on modifying network structures, with limited progress in extracting and describing features of RNA sequences and structures. In this study, we propose a novel two-dimensional Kmer cross-encoding approach based on an improved traditional Kmer encoding to predict miRNA-lncRNA interactions. This encoding integrates the features of miRNA and lncRNA into a meaningful encoded image, allowing for interactive interpretation. Furthermore, it combines neural networks that process sequence and image information. The proposed method named PmliGKKS was trained and tested on species from four different species, with independent testing conducted on two additional species. The results obtained using our approach demonstrate significant improvements compared to several state-of-the-art methods.

ARES-Kcr: A New Network Model Utilizing Attention Mechanism and Residual Structure for the Prediction of Lysine Crotonylation Sites.

Lysine crotonylation (Kcr), as a significant post-translational modification of protein, exists in the core histones and some non histones of many organisms, and plays a crucial regulatory role in many biological processes such as gene expression, cell development, and disease treatment. Due to the high cost, time-consuming and labor-intensive nature of traditional biological experimental methods, it is necessary to develop efficient, low-cost and accurate calculation methods for identifying crotonylation sites. Therefore, we propose a new network model called ARES-Kcr, which extracts three types of features from different perspectives and integrates convolutional modules, attention mechanisms, and residual modules for feature fusion to improve prediction ability in this paper. Our model performs significantly better than other models on the benchmark dataset, with an average AUC of 92% in the independent test set, demonstrating its excellent predictive ability.

4mCFSNet: A Feature Fusion-Based Multi-Scale Representations for Predicting DNA N4-Methylcytosine Sites.

N4-methylcytosine (4mC) is a very important epigenetic modification that regulates DNA expression, repair and replication. Traditional experimental methods for 4mC site detection are both time consuming and laborious. Therefore, the development of computational methods is necessary. But mining the internal information of DNA sequences remains a great challenge. In this paper, we propose a novel 4mC deep learning prediction method, named 4mCFSNet. Firstly, we encode the sequences using one-hot. Secondly, we construct multi-scale fusion modules to fully extract biological sequence information by overlapping multi-scale channel input features. Finally, we use fully connected layers and class weights for multi-species classification prediction. The average MCC of our proposed method on six species is about 2% higher than the optimal method, and the average ACC is about 1% higher.