Special Issue 2021
Call for Paper for the Special Issue of Molecular Diversity on
“AI and ML for Small Molecule Drug Discovery in the Big Data Era”
Advancements in the informatics and omics based technologies have enhanced our ability to generate data at lower costs. The recent emergence of ‘big data’ in chemistry and biology has fundamentally revolutionized molecular biology and drug development paradigms. The recent availability of open data though various databases and online resources has led to simply too much information (big data) for a human being to assimilate using traditional research methods. The emergence of machine learning (ML) and artificial intelligence (AI) offers guidance to the research scientists to process, analyze and understand the data, and their extensive application appears to be the future for drug discovery.
The traditional drug discovery process is very costly and lengthy with limited success probability. The chemical space is very large (an estimated number of small organic molecules is 1060), a fraction of which we have explored. AI can be used to explore the chemical space and to understand the pattern of the complex big data. AI algorithms can make accurate predictions about complex systems involving the vast and unexplored space of molecules, reactions, and biological interactions. ML techniques have potential to identify new drug candidates in much less time than the conventional research. However, the field is still young, having ample scope for improvements in the accuracy of AI algorithms and the adoption of more standardized and rigorous benchmarks so that the discipline can mature and improve further.
Supervised learning including regression and classification based modeling (for example, logistic regression, random forest, naïve Bayes, support vector machine, etc.) and unsupervised learning methods (for example, clustering methods, dimensionality reduction, self-organizing map, etc.), have been used in the literature for drug discovery problems. Other methods such as semisupervised learning, active learning, transfer learning and multitask learning are also in use. Bayesian algorithms, instance based methods, decision trees, ensemble based methods, principal component analysis, artificial neural network, deep neural network, etc. are some of the popular algorithms used for cheminformatic modeling with diverse applications like target identification and validation, structure-based virtual screening, ligand-based virtual screening, de novo drug design, ADMET predictions, computational drug repurposing, etc.
Molecular Diversity (EISSN 1573-501X) of Springer Nature (https://www.springer.com/journal/11030) with the current Impact factor 2.013 (2019) will publish a special issue on “AI and ML for Small Molecule Drug Discovery in the Big Data Era” in 2021 to showcase the latest developments in this field. Researchers working in this fascinating area are welcome to submit their fine work for consideration of inclusion in this special issue.
Topic to be covered but not limited to
1. Different AI and ML methods in drug discovery - Artificial neural network, Deep learning, Generative Adversarial Networks, Support vector machine, Random forest, Bayesian Algorithms, Instance-Based Methods, Decision trees, Ensemble methods, Dimensionality reduction, AI-based QSAR, Pattern recognition, Unsupervised learning methods, etc.
2. Applications of AI and ML in target identification and validation
3. Applications of AI and ML in structure-based virtual screening
4. Applications of AI and ML in ligand-based virtual screening
5. Applications of AI and ML in de novo drug design
6. Applications of AI and ML in ADMET predictions
7. Applications of AI and ML in computational drug repurposing
8. ML-based biomarker discovery and drug sensitivity predictive models
9. Deep learning in drug discovery
10. Multitask learning in drug discovery
11. Applications of AI and ML methods in development of drugs against diseases like malaria, tuberculosis, viral infections, neglected tropical & rare diseases, etc.
12. Applications of AI and ML in computational pathology
Projected Publication Timeline Volume 25, Issue 3, 2021
Manuscript Due March 31, 2021
Revision Due May 31, 2021
Issue Published August 2021
For any query regarding this special issue, please contact
Prof. Kunal Roy, firstname.lastname@example.org