From Black Boxes to Actionable Insights: A Perspective on Explainable Artificial Intelligence for Scientific Discovery.

The application of Explainable Artificial Intelligence (XAI) in the field of chemistry has garnered growing interest for its potential to justify the prediction of black-box machine learning models and provide actionable insights. We first survey a range of XAI techniques adapted for chemical applications and categorize them based on the technical details of each methodology. We then present a few case studies to illustrate the practical utility of XAI, such as identifying carcinogenic molecules and guiding molecular optimizations, in order to provide chemists with concrete examples of ways to take full advantage of XAI-augmented machine learning for chemistry. Despite the initial success of XAI in chemistry, we still face the challenges of developing more reliable explanations, assuring robustness against adversarial actions, and customizing the explanation for different applications and needs of the diverse scientific community. Finally, we discuss the emerging role of large language models like GPT in generating natural language explanations and discusses the specific challenges associated with them. We advocate that addressing the aforementioned challenges and actively embracing new techniques may contribute to establishing machine learning as an indispensable technique for chemistry in this digital era.

CODD-Pred: A Web Server for Efficient Target Identification and Bioactivity Prediction of Small Molecules.

Target identification and bioactivity prediction are critical steps in the drug discovery process. Here we introduce CODD-Pred (COmprehensive Drug Design Predictor), an online web server with well-curated data sets from the GOSTAR database, which is designed with a dual purpose of predicting potential protein drug targets and computing bioactivity values of small molecules. We first designed a double molecular graph perception (DMGP) framework for target prediction based on a large library of 646 498 small molecules interacting with 640 human targets. The framework achieved a top-5 accuracy of over 80% for hitting at least one target on both external validation sets. Additionally, its performance on the external validation set comprising 200 molecules surpassed that of four existing target prediction servers. Second, we collected 56 targets closely related to the occurrence and development of cancer, metabolic diseases, and inflammatory immune diseases and developed a multi-model self-validation activity prediction (MSAP) framework that enables accurate bioactivity quantification predictions for small-molecule ligands of these 56 targets. CODD-Pred is a handy tool for rapid evaluation and optimization of small molecules with specific target activity. CODD-Pred is freely accessible at http://codd.iddd.group/.

One-step hydrothermal synthesis of nitrogen-doped carbon dots for high-sensitivity visual detection of nitrite and ascorbic acid.

Ascorbic acid (AA) is an indispensable vitamin for the human body and is associated with critical processes of human metabolism. However, excessive intake of AA can also have a negative impact on human health. Nitrite is a commonly used food additive, and its overdose can increase the risk of cancer. Therefore, the detection of nitrite and vitamins is generally recognized to be meaningful. In this study, red-fluorescence and yellow-fluorescence CDs (r-CDs/y-CDs) were synthesized by a one-step hydrothermal method using o-phenylenediamine as the only carbon source. These two types of CDs exhibited good detection accuracy, detection limit and selectivity towards nitrite and AA (the detection limits are 0.47 µM and 45.1 µM, respectively). The long wavelength luminescent CDs prepared in this experiment also have high quantum yield (QY), which is of great significance to the visual detection effect. Under weak acidic conditions, the amino group on the surface of r-CDs can coordinate with nitrite and react to generate diazo groups, leading to the fluorescence quenching of CDs. The coordination between the y-CDs and the amino group on the surface of AA connects the adjacent y-CDs to form aggregates, which increases the non-radiative transition of electrons and induces the fluorescence quenching of CDs. This study proposes a new idea for the preparation of carbon dots for the determination of NO2- and AA in solutions, which expands the application of fluorescent CD detection.

Preparation of N/S doped carbon dots and their application in nitrite detection.

Detection of carcinogens is generally recognized to be meaningful, especially for nitrites (NO2 -). Here blue-green fluorescent carbon dots (CDs) were successfully synthesized by using p-aminobenzenesulfonic acid, and their surfaces were identified to be abundant in the functional groups of amino, hydroxyl, and sulfuric acid. Importantly, the sulfuric acid group and aromatic primary ammonia groups on the surfaces of CDs showed the interactions with the nitrites to cause fluorescence quenching. The novel CDs showed high sensitivity and selectivity for NO2 - detection with a low detection limit of 0.03 mM in water due to the fluorescence quenching effect of the CDs. Consequently, the proposed CDs here may provide a new way of monitoring NO2 - in the target samples.