Molecular fingerprints are strong models for peptide function prediction

Machine learning assisted QSPR model for prediction of ionic liquid’s refractive index and viscosity: The effect of representations of ionic liquid and ensemble model development

Rule extraction based on PROMETHEE-assisted multi-objective genetic algorithm for generating interpretable neural networks

Quantitative structure-activity relationship (QSAR) models based on machine learning algorithms are powerful tools to expedite drug discovery processes and therapeutics development. Given the cost in acquiring large-sized training datasets, it is useful to examine if QSAR analysis can reasonably predict drug activity with only a small-sized dataset (size < 100) and benchmark these small-dataset QSAR models in application-specific studies. To this end, here we present a systematic benchmarking study on small-dataset QSAR models built for prediction of effective Wnt signaling inhibitors, which are essential to therapeutics development in prevalent human diseases (e.g., cancer). Specifically, we examined a total of 72 two-dimensional (2D) QSAR models based on 4 best-performing algorithms, 6 commonly used molecular fingerprints, and 3 typical fingerprint lengths. We trained these models using a training dataset (56 compounds), benchmarked their performance on 4 figures-of-merit (FOMs), and examined their prediction accuracy using an external validation dataset (14 compounds). Our data show that the model performance is maximized when: 1) molecular fingerprints are selected to provide sufficient, unique, and not overly detailed representations of the chemical structures of drug compounds; 2) algorithms are selected to reduce the number of false predictions due to class imbalance in the dataset; and 3) models are selected to reach balanced performance on all 4 FOMs. These results may provide general guidelines in developing high-performance small-dataset QSAR models for drug activity prediction.

It has been shown that Variable Bit Rate (VBR) video exhibits long-range dependence characteristics. Several models have been proposed to synthesize traces whose autocorrelation match that of the video traffic traces. Given the relationship between self-similarity and long- range dependence, we investigate the application of discrete-time linear scale invariant systems provided by Zhao and Rao to modeling VBR video traffic. This formulation, called a DLSI system, was derived using a continuous dilation operator in discrete-time as a direct analog of the continuous-time linear scale-invariant system formulation of Wornell and differs from other approaches such as wavelet based construction. While simulations had shown that DLSI systems were capable of generating self- similar data such as those found in network traffic, questions regarding their long-range behavior remained. Answers to these questions are dealt in this paper. We present an alternative modeling technique to model VBR video traffic using DLSI system. The proposed model has a fractional pole-zero structure and provides a good model for long-range dependent and self-similar time series. The behavioral pattern of the fractional pole-zero filter for a range of the Hurst parameter is also shown. The time-domain characteristics of the filter are analyzed using power series approximations. Comparison of output autocorrelation function (ACF) to those of the video trace ACF is made. We conclude that some samples of video data are indeed better modeled by the LSI models than the conventional models such as Markovian and Long-range dependent models.

We analyze 20 large sets of actual variable-bit-rate (VBR) video data, generated by a variety of different codecs and representing a wide range of different scenes. Performing extensive statistical and graphical tests, our main conclusion is that long-range dependence is an inherent feature of VBR video traffic, i.e., a feature that is independent of scene (e.g., video phone, video conference, motion picture video) and codec. In particular, we show that the long-range dependence property allows us to clearly distinguish between our measured data and traffic generated by VBR source models currently used in the literature. These findings give rise to novel and challenging problems in traffic engineering for high-speed networks and open up new areas of research in queueing and performance analysis involving long-range dependent traffic models. A small number of analytic queueing results already exist, and we discuss their implications for network design and network control strategies in the presence of long-range dependent traffic. >

Phase unwrapping is one of the key problems in fringe projection profilometry (FPP). These years, CNN-based and Transformer-based models are widely used in FPP phase unwrapping. Unfortunately, the limitation of CNNs in long-range modeling capabilities prevent them from effectively extracting fine-grained features in wrapping phase images, while Transformer-based models struggle with efficiently handling long-range dependencies due to their local focus or computational demands. Recent studies demonstrate that Mamba, a novel selective state space model (SSM), achieves efficient modeling of long-range dependencies by dynamically adapting its parameters based on input context, particularly in tasks such as long-sequence modeling. Inspired by this, we propose a hybrid deep learning model integrating Res-UNet with Mamba (CNN-Mamba network, CMNet) for phase unwrapping of single-frame fringe patterns. First, a new dual-branch skip connection module based on Mamba is proposed named convolution weighted feature fusion SSM (Conv_WFF-SSM) is proposed, integrating Mamba’s selective state space mechanisms with multi-scale convolutional feature weighting to simultaneously address long-range interaction modeling and hierarchical feature preservation during phase expansion. Second, a parameter-free attention module (PFAM) is introduced into the encoder and decoder to reduce information loss caused by downsampling and upsampling operations without increasing network parameters. Finally, the iterative attentional feature fusion (IAFF) module integrated into residual block to instead of the ordinary sum operation for the first time. Experiments demonstrate the validity and robustness of the proposed technique.

Remote sensing image dehazing is a well-known remote sensing image processing task focused on restoring clean images from hazy images. The Transformer network, based on the self-attention mechanism, has demonstrated remarkable advantages in various image restoration tasks, due to its capacity to capture long-range dependencies within images. However, it is weak at modeling local context. Conversely, convolutional neural networks (CNNs) are adept at capturing local contextual information. Local contextual information could provide more details, while long-range dependencies capture global structure information. The combination of long-range dependencies and local context modeling is beneficial for remote sensing image dehazing. Therefore, in this paper, we propose a CNN-based adaptive local context enrichment module (ALCEM) to extract contextual information within local regions. Subsequently, we integrate our proposed ALCEM into the multi-head self-attention and feed-forward network of the Transformer, constructing a novel locally enhanced attention (LEA) and a local continuous-enhancement feed-forward network (LCFN). The LEA utilizes the ALCEM to inject local context information that is complementary to the long-range relationship modeled by multi-head self-attention, which is beneficial to removing haze and restoring details. The LCFN extracts multi-scale spatial information and selectively fuses them by the the ALCEM, which supplements more informative information compared with existing regular feed-forward networks with only position-specific information flow. Powered by the LEA and LCFN, a novel Transformer-based dehazing network termed LCEFormer is proposed to restore clear images from hazy remote sensing images, which combines the advantages of CNN and Transformer. Experiments conducted on three distinct datasets, namely DHID, ERICE, and RSID, demonstrate that our proposed LCEFormer achieves the state-of-the-art performance in hazy scenes. Specifically, our LCEFormer outperforms DCIL by 0.78 dB and 0.018 for PSNR and SSIM on the DHID dataset.

Multimodal medical image synthesis plays a crucial supportive role in research within the field of biomechanics, providing high-precision data and analytical methods for studies on anatomical structures, tissue characteristics, and mechanical modeling. However, due to practical constraints, certain modalities of medical images may be difficult to obtain, posing challenges for model training and high-accuracy biomechanical research. Existing methods employ convolutional neural network (CNN)-based generative adversarial models to synthesize missing modality information across modalities. However, CNNs are limited in their ability to model long-range dependencies. Transformers offer a new paradigm to address these limitations, yet their high computational and memory demands remain a significant drawback. To tackle these challenges, we propose a novel generative adversarial model, termed the Convolutional Attention Latent Feature GAN (CALF-GAN), which leverages multi-scale convolutional attention for cross-modal medical image synthesis. A dedicated latent attribute separation module is employed to disentangle modality-specific features between source and target modality images, enhancing the synthesis of medical semantics, such as pixel intensity values. Furthermore, to improve the model’s capacity for long-range dependency modeling while reducing computational overhead, we design a generation module based on multi-scale convolutional attention, capturing long-range dependencies using only convolutional operations. Extensive experiments conducted on various medical image datasets demonstrate that CALF-GAN achieves remarkable generalizability and outstanding overall performance under low memory requirements, making it well-suited for application in high-precision biomechanics research.

PurposeDrug-induced liver injury (DILI) is a significant concern in drug development, often leading to the discontinuation of clinical trials and the withdrawal of drugs from the market. This study explores the application of graph neural networks (GNNs) for DILI prediction, using molecular graph representations as the primary input.MethodsWe evaluated several GNN architectures, including Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), Graph Sample and Aggregation (GraphSAGE), and Graph Isomorphism Networks (GINs), using the latest FDA DILI dataset and other molecular property prediction datasets. We introduce a novel approach that creates a custom graph dataset, driven by molecular optimisation, that incorporates detailed and realistic chemical features such as bond lengths and partial charges as input into the GNN models. We have named our model approach DILIGeNN.ResultsDILIGeNN achieved an AUC of 0.897 on the DILI dataset, surpassing the current state-of-the-art model in the DILI prediction task. Furthermore, DILIGeNN outperformed the state-of-the-art in other graph-based molecular prediction tasks, achieving an AUC of 0.918 on the Clintox dataset, 0.993 on the BBBP dataset, and 0.953 on the BACE dataset, indicating strong generalisation and performance across different datasets.ConclusionDILIGeNN, utilising a single graph representation as input, outperforms the state-of-the-art methods in DILI prediction that incorporate both molecular fingerprint and graph-structured data. These findings highlight the effectiveness of our molecular graph generation and the GNN training approach as a powerful tool for early-stage drug development and drug repurposing pipeline.Scientific Contribution: DILIGeNN is a GNN framework that extracts graph features from 3D optimised molecular structures as is done in target-based drug discovery and molecular docking simulation. Our method is the first to encode spatial and electrostatic information into a single graph representation, as opposed to other work that require multiple graphs or additional chemical descriptors for feature representation. Our approach, using warm starts following repeated early stopping during training, outperforms the current state-of-the-art methods in liver toxicity (DILI), permeability (BBBP) and activity (BACE) prediction tasks.Graphic

Neural substrates of decision-making

Recent researches have highlighted the long-range dependence and impulsive nature of the aggregated network traffic, and consequently long-range dependent α-stable processes model has been widely used to capture the network traffic. This paper presents an analytical demonstration of aggregated network traffic converging to long-range dependent α-stable processes, which provides inherent mechanism explanation for the widely used long-range dependent α-stable processes model. Simulations validate the accuracy of the presented theoretical analysis.

Multifractional and long-range dependent characteristics for remaining useful life prediction of cracking gas compressor

In medical image segmentation, U-Net and its variants are widely utilized. However, CNN-based models are limited in long-range dependency modeling, while Transformer-based models are constrained by quadratic computational complexity. Recently, state space models (SSMs) such as Mamba have emerged as a promising solution, as they excel at modeling long-range interactions while maintaining linear computational complexity. This paper proposes the CIHM, a hybrid Mamba model that provides contextual insights. Its key component is the context-insight Mamba-CNN layer (CiMC layer). The CiMC layer adopts a hybrid Mamba-CNN approach to process features across different scales and levels in parallel, thereby enhancing the semantic segmentation of medical images. Specifically, the SSM branch of this layer captures long-range dependencies with comprehensive contextual semantic understanding, while the CNN block achieves superior local feature extraction using T-shaped convolution that focuses on patch centers. Moreover, to further aggregate multi-scale contextual information of medical images, we design the multiscale refining detail bridge (MRDB) module. This module significantly improves feature representation through the dense multiplication and concatenation of features for fusion. Comprehensive experiments on the ISIC2017, ISIC2018, Synapse, and DSB18 datasets demonstrate that CIHM exhibits strong competitive performance and efficiency in medical image segmentation tasks. Notably, CIHM achieves better segmentation performance than the well-known U-Net and the modern hybrid network U-Mamba, while reducing the parameter count by 40 times and 345 times, respectively. This demonstrates the potential of our approach to advance model lightweighting.

Machine learning (ML) is increasingly being used to guide drug discovery processes. When applying ML approaches to chemical datasets, molecular descriptors and fingerprints are typically used to represent compounds as numerical vectors. However, in recent years, end-to-end deep learning (DL) methods that can learn feature representations directly from line notations or molecular graphs have been proposed as alternatives to using precomputed features. This study set out to investigate which compound representation methods are the most suitable for drug sensitivity prediction in cancer cell lines. Twelve different representations were benchmarked on 5 compound screening datasets, using DeepMol, a new chemoinformatics package developed by our research group, to perform these analyses. The results of this study show that the predictive performance of end-to-end DL models is comparable to, and at times surpasses, that of models trained on molecular fingerprints, even when less training data is available. This study also found that combining several compound representation methods into an ensemble can improve performance. Finally, we show that a post hoc feature attribution method can boost the explainability of the DL models.

Accelerated prediction of molecular properties for per- and polyfluoroalkyl substances using graph neural networks with adjacency-free message passing.