AI for Drug Discovery / Machine Learning | Immunocure Drug Discovery| AI / ML CRO for Drug Discovery

Small-molecule drug discovery SMDD is a multidimensional challenge that involves huge expenditure and time-consuming. The average time required from the invention to market is about 14 years and cost around US$2.8 billion. Two major factors for the failures in discovery are efficacy and toxicity. Though the discovery of drugs is a slow process with high investment, the pharmaceutical companies and academic institutes are still spending money because of high commercial potential and also benefits society. A huge amount of experimental data accumulated from the past decades including invitro (biochemical) assays, in vivo assays and clinical trials. This data will become a valuable source for learning and understanding the success and failure of compounds in the entire discovery process. The acquired knowledge will be useful to predict future drug candidates in the discovery experiments. The generation of the knowledge/hypothesis from the known data can be implemented by using machine learning/deep learning methods. Since the data acuminated is huge (Big-Data), proper curation, efficient mining and building hypothesis (ML/DL models) need to be implemented in the drug discovery and development pipeline of the pharmaceutical industry to increase their success rates. The integration of these tools as Artificial Intelligence (AI) can serve end-to-end drug discovery and development. Thus, combining the drug discovery process with AI transform the paradigm of drug discovery.
Millions of experimental data (known data) available in the public domain (PubChem, CheMBL, Binding DB, PDB, etc.). The experimental data includes in-vitro and in vivo data for each disease, ADME/Tox and many more. Properly curated data should be considered for the generation of precise ML/DL models. Two types of machine learning methods are widely used in the small molecule drug discovery, unsupervised and supervised. Unsupervised methods are used to cluster the molecules based on chemical similarity. As the data is large, quicker clustering methods, k-means, k-median, mean-shifting, Gaussian mixture can be applied to yield better results. The clustering methods are useful to identify the nearest neighbors and have a greater application in repurposing and off-targets prediction. Supervised machine learning models are useful in generating the models for the data sets having the experimental activity. These methods are predictive methods with either quantitative/continue or qualitative/categorical based on the experimental activity of the training data. Generating the models for each protein or type of disease and when applied, will classify the unexplored data more precisely to identify new hits. However, precision mainly depends on the quality of the input data. The predictive methods combining with molecular modeling will accelerate the discovery process from hit identification to lead optimization.
Part of this article is missing due to restrictions. Read the full article here immunocure.us/...
Recently deep neural network (DNN) gain importance not only in drug discovery also in other areas of science and business. DNN is a deep learning neural network method that builds hierarchical internal representations of the input data with the help of multiple hidden layers. Four major DNNs, Convolutional neural networks (CNN), Recurrent neural network (RNN), Deep autoencoders (AE), Deep belief network (DBN) are having their own advantage for the model generation. These methods are applied in the prediction of biological activity, ADMET properties, and physicochemical parameters. For small training data, ML methods will perform equally or better than DNN, but with large datasets, DNN will outperform ML. Overfitting is a major challenge during model generation, recent developments available to overcome this challenge such as DropOut and DropConnect. A significant development in these methods in the areas of de novo design, the binding energy between ligand-receptor, chemical syntheses, nanoparticles, formulations and many more.
Visit our website: immunocure.us
Contact us for presentation or discussion: [e-mail]
Our social media handles:
LinkedIn: / immunocureds
Facebook: / immunocureds
Twitter: / immunocureds
Reach Us Here: immunocure.us/...
Our contact details:
USA - PARENT OFFICE
Immunocure LLC, 94 Greene Ave # A, Brooklyn, NY 11238, United States
[phone]
[phone]
INDIA - SUBSIDIARY
Immunocure Discovery Solutions Pvt Ltd, Life Science Incubator, ICICI Knowledge Park, Genome Valley,
Shamir Pet
Hyderabad
Telangana, India - 500101
[phone]
This presentation contains images that were used under a Creative Commons License. Click here to see the full list of images and attributions:
link.attribute...