Molecular language models: RNNs or transformer?

Language models have shown the capacity to learn complex molecular distributions. In the field of molecular generation, they are designed to explore the distribution of molecules, and previous studies have demonstrated their ability to learn molecule sequences. In the early times, recurrent neural networks (RNNs) were widely used for feature extraction from sequence data and have been used for various molecule generation tasks. In recent years, the attention mechanism for sequence data has become popular. It captures the underlying relationships between words and is widely applied to language models. The Transformer-Layer, a model based on a self-attentive mechanism, also shines the same as the RNN-based model. In this research, we investigated the difference between RNNs and the Transformer-Layer to learn a more complex distribution of molecules. For this purpose, we experimented with three different generative tasks: the distributions of molecules with elevated scores of penalized LogP, multimodal distributions of molecules and the largest molecules in PubChem. We evaluated the models on molecular properties, basic metrics, Tanimoto similarity, etc. In addition, we applied two different representations of the molecule, SMILES and SELFIES. The results show that the two language models can learn complex molecular distributions and SMILES-based representation has better performance than SELFIES. The choice between RNNs and the Transformer-Layer needs to be based on the characteristics of dataset. RNNs work better on data focus on local features and decreases with multidistribution data, while the Transformer-Layer is more suitable when meeting molecular with larger weights and focusing on global features.

Path reliability-based graph attention networks.

Self-attention mechanism has been successfully introduced in Graph Neural Networks (GNNs) for graph representation learning and achieved state-of-the-art performances in tasks such as node classification and node attacks. In most existing attention-based GNNs, attention score is only computed between two directly connected nodes with their representation at a single layer. However, this attention score computation method cannot account for its multi-hop neighbors, which supply graph structure information and have influence on many tasks such as link prediction, knowledge graph completion, and adversarial attack as well. In order to address this problem, in this paper, we propose Path Reliability-based Graph Attention Networks (PRGATs), a novel method to incorporate multi-hop neighboring context into attention score computation, enabling to capture longer-range dependencies and large-scale structural information within a single layer. Moreover, path reliability-based attention layer, a core layer of PRGATs, uses a resource-constrain allocation algorithm to compute the reliable path and its attention scores from neighboring nodes to non-neighboring nodes, increasing the receptive field for every message-passing layer. Experimental results on real-world datasets show that, as compared with baselines, our model outperforms existing methods up to 3% on standard node classification and 12% on graph universal adversarial attack.

Smartphone recognition-based immune microparticles for rapid on-site visual data-sharing detection of Newcastle disease virus.

Since the last century, animal viruses have posed great threats to the health of humans and the farming industry. Therefore, virus control is of great urgency, and regular, timely, and accurate detection is essential to it. Here, we designed a rapid on-site visual data-sharing detection method for the Newcastle disease virus with smartphone recognition-based immune microparticles. The detection method we developed includes three major modules: preparation of virus detection vectors, sample detection, and smartphone image analysis with data upload. First, the hydrogel microparticles containing active carboxyl were manufactured, which coated nucleocapsid protein of NDV. Then, HRP enzyme-labeled anti-NP nanobody was used to compete with the NDV antibody in the serum for color reaction. Then the rough detection results were visible to the human eyes according to the different shades of color of the hydrogel microparticles. Next, the smartphone application was used to analyze the image to determine the accurate detection results according to the gray value of the hydrogel microparticles. Meanwhile, the result was automatically uploaded to the homemade cloud system. The total detection time was less than 50 min, even without trained personnel, which is shorter than conventional detection methods. According to experimental results, this detection method has high sensitivity and accuracy. And especially, it uploads the detection information via a cloud platform to realize data sharing, which plays an early warning function. We anticipate that this rapid on-site visual data-sharing detection method can promote the development of virus self-checking at home.