The field of catalysis holds paramount importance in shaping the trajectory of sustainable development, prompting intensive research efforts to leverage artificial intelligence (AI) in catalyst design. Presently, the fine-tuning of open-source large language models (LLMs) has yielded significant breakthroughs across various domains such as biology and healthcare. Drawing inspiration from these advancements, we introduce CataLM (Catalytic Language Model), a large language model tailored to the domain of electrocatalytic materials. Our findings demonstrate that CataLM exhibits remarkable potential for facilitating human-AI collaboration in catalyst knowledge exploration and design. To the best of our knowledge, CataLM stands as the pioneering LLM dedicated to the catalyst domain, offering novel avenues for catalyst discovery and development.

Motivation: Extensive efforts have been devoted to understanding the antigenic peptides binding to MHC class I and II molecules since they play a fundamental role in controlling immune responses and due their involvement in vaccination, transplantation, and autoimmunity. The genes coding for the MHC molecules are highly polymorphic, and it is difficult to build computational models for MHC molecules with few know binders. On the other hand, previous studies demonstrated that some MHC molecules share overlapping peptide binding repertoires and attempted to group them into supertypes. Herein, we present a framework of the utility of supertype clustering to gain more information about the data to improve the prediction accuracy of class II MHC-peptide binding. Results: We developed a new method, called superMHC, for class II MHC-peptide binding prediction, including three MHC isotypes of HLA-DR, HLA-DP, and HLA-DQ, by using supertype clustering in conjunction with RLS regression. The supertypes were identified by using a novel repertoire dissimilarity index to quantify the difference in MHC binding specificities. The superMHC method achieves the state-of-the-art performance and is demonstrated to predict binding affinities to a series of MHC molecules with few binders accurately. These results have implications for understanding receptor-ligand interactions involved in MHC-peptide binding.

RNA-protein interactions (RPIs) have critical roles in numerous fundamental biological processes, such as post-transcriptional gene regulation, viral assembly, cellular defence and protein synthesis. As the number of available RNA-protein binding experimental data has increased rapidly due to high-throughput sequencing methods, it is now possible to measure and understand RNA-protein interactions by computational methods. In this study, we integrate a sequence-based derived kernel with regularized least squares to perform prediction. The derived kernel exploits the contextual information around an amino acid or a nucleic acid as well as the repetitive conserved motif information. We propose a novel machine learning method, called RPiRLS to predict the interaction between any RNA and protein of known sequences. For the RPiRLS classifier, each protein sequence comprises up to 20 diverse amino acids but for the RPiRLS-7G classifier, each protein sequence is represented by using 7-letter reduced alphabets based on their physiochemical properties. We evaluated both methods on a number of benchmark data sets and compared their performances with two newly developed and state-of-the-art methods, RPI-Pred and IPMiner. On the non-redundant benchmark test sets extracted from the PRIDB, the RPiRLS method outperformed RPI-Pred and IPMiner in terms of accuracy, specificity and sensitivity. Further, RPiRLS achieved an accuracy of 92% on the prediction of lncRNA-protein interactions. The proposed method can also be extended to construct RNA-protein interaction networks. The RPiRLS web server is freely available at http://bmc.med.stu.edu.cn/RPiRLS.

China is experiencing severe PM_2.5 (fine particles with a diameter of 2.5 μg or smaller) pollution problem. Little is known, however, about how the increasing concentration trend is spatially distributed, nor whether there are some areas that experience a stable or decreasing concentration trend. Managers and policymakers require such information to make strategic decisions and monitor progress towards management objectives. Here, we present a pixel-based linear trend analysis of annual PM_2.5 concentration variation in China during the period 1999–2016, and our results provide guidance about where to prioritize management efforts and affirm the importance of controlling coal energy consumption. We show that 87.9% of the whole China area had an increasing trend. The drastic increasing trends of PM_2.5 concentration during the last 18 years in the Beijing–Tianjin–Hebei region, Shandong province, and the Three Northeastern Provinces are discussed. Furthermore, by exploring regional PM_2.5 pollution, we find that Tarim Basin endures a high PM_2.5 concentration, and this should have some relationship with oil exploration. The relationship between PM_2.5 pollution and energy consumption is also discussed. Not only energy structure reconstruction should be repeatedly emphasized, the amount of coal burned should be strictly controlled.

With the development of natural language processing techniques(NLP), automatic diagnosis of eye diseases using ophthalmology electronic medical records (OEMR) has become possible. It aims to evaluate the condition of both eyes of a patient respectively, and we formulate it as a particular multi-label classification task in this paper. Although there are a few related studies in other diseases, automatic diagnosis of eye diseases exhibits unique characteristics. First, descriptions of both eyes are mixed up in OEMR documents, with both free text and templated asymptomatic descriptions, resulting in sparsity and clutter of information. Second, OEMR documents contain multiple parts of descriptions and have long document lengths. Third, it is critical to provide explainability to the disease diagnosis model. To overcome those challenges, we present an effective automatic eye disease diagnosis framework, NEEDED. In this framework, a preprocessing module is integrated to improve the density and quality of information. Then, we design a hierarchical transformer structure for learning the contextualized representations of each sentence in the OEMR document. For the diagnosis part, we propose an attention-based predictor that enables traceable diagnosis by obtaining disease-specific information. Experiments on the real dataset and comparison with several baseline models show the advantage and explainability of our framework.