In our med-term business plan Shionogi Growth Strategy 2020 (SGS2020), we've declared we'll grow as a drug discovery-based pharmaceutical company with global society. Now, we're promoting drug discovery research to supply the best possible medicine to protect the health and wellbeing of the global patients. Our source of the competitiveness is a small molecule drug discovery which has produced several launched drugs during the past 30 years. To supply innovative medicines in continuous way, we need a drug discovery platform consists of research assets that could make it possible to produce a new drug candidate. We've successfully achieved to establish such platform, especially in the infectious disease area, that has produced several β-lactam antibiotics and anti-viral drugs represented by Tivicay and Zofluza. Our next challenge is to improve our productivity in dramatically, and we're promoting a research not only for repeated improvements of the small molecule drug discovery, but also expanding our chemistry strength from the small molecules to the medium-sized molecules.

Recently AI（Artificial Intelligence）has begun to penetrate our lives as long as we do not see the topic of AI in the media etc. As a technology of the Fourth Industrial Revolution, AI is definitely changing our lives. In November 2016, we established “Life Intelligence Consortium（LINC）” with the aim of developing AI in the fields of pharmaceutical science, chemistry, food, medical care and healthcare related life science. LINC is developing more than 30 AIs covering a wide range of the entire process of drug development from target exploration to clinical trials. This special issue focuses on “drug target molecule search, protein structure prediction, ADMET prediction, molecular design AI, synthesis path prediction” which is a field closely related to medicinal chemistry among drug development processes, and we will introduce their latest AI technology.

The amount of data that can be collected and processed has increased dramatically due to factors including the wider availability of devices to provide comprehensive measurement of biomolecules, the migration from paper to electronic medical records, and the rapid progress of information technology. Meanwhile, the pharmaceutical industry faces problems such as an increase in the research and development cost and the depletion of new drug targets. For these reasons, attempts to utilize Artificial Intelligence (AI) for various kinds of “big data” in drug discovery research have attracted attention in recent years. Among the AI technologies, there are great expectations for deep learning in particular; it has been used in various applications such as predicting disease by using electronic medical records, or searching for disease-related factors by integrating omics data. Since it is expected that the amount of data available keeps increasing in the future, further development of AI technologies and infrastructure for data collection are required in order to make effective use of data.

Since October 2015, the drug discovery informatics project has been started at the Japan Agency for Medical Research and Development (AMED). We are building an informatics platform to predict ADMET, which is a bottle neck of drug discovery, and design promising drug candidates. On the other hand, WG 5 of LINC is developing a structure generation AI and a synthetic path AI in addition to prediction by ADMET AI, and aims at a structure proposal AI that enables simultaneous optimization by integrating them. In this chapter, we review the recent situation of ADMET prediction including AI development.

Life INtelligent Consortium (LINC), where life-related companies, IT companies and academia are making collaborative efforts, is a consortium aiming for a process innovation of drug discovery, development, and life-cycle management based on artificial intelligence (AI) technologies. LINC working group 4 (WG4) consists of four projects; Project 11 (PJ11)–Prediction of 3D structure of proteins, PJ12–Prediction of a binding of ligand to proteins, PJ13–Integration of MD simulations and AI, and PJ14–Development of AI force field. Recent applications of AI technologies, especially of deep learning (DL) applied to the areas corresponding to the four projects of WG4, are reported firstly so as to help understanding of the technology trends in this field. Then, the activities for each project of WG4 are briefly introduced.

Despite the fact that organic molecules are composed of only a few kinds of atoms on the basis of carbon and hydrogen atoms, molecules having different functional groups can be infinitely designed as a result of permutation and combination of constituent atoms. When synthesizing them in practice, first do thought experiments called retrosynthesis, continue dividing so that the finished puzzles are returned to pieces, and design a synthetic route from available raw materials. Whether the synthesis according to this route will actually work or not depends on the quality and accuracy of the first thought experiment. In this retrosynthesis, since the use of all past organic reaction data is the key, the expectation that it can be performed at a higher degree by the method using AI has increased, and great progress has been made in recent years.

To modify endogenous protein selectively under biological conditions, ligand-directed (LD) chemistry has recently attracted much attention. LD chemistry is a novel affinity labeling method utilizing a reactive moiety that is cleaved upon chemical modification, allowing chemical labeling of a target protein without loss of function. Moreover, a few of LD chemistry have succeeded in selective labeling of specific protein not only in living cells but also in more complex conditions, such as live brain tissues or living mice. LD chemistry can be applied to the construction of biosensor for drug screening and the development of covalent inhibitors in addition to bioimaging. This review describes the outline of LD chemistry and its application for drug discovery.

54^th International Conference on Medicinal Chemistry (Rencontres Internationales de Chimie Thérapeutique: RICT 2018) was held in Strasbourg, France on July 4-6, 2018. The theme of this meeting was “Interfacing Chemical Biology and Drug Discovery”. 23 plenary lectures, 6 keynote lectures, and 214 poster presentations were given. In this report, outlines of the meeting and some lectures are described.