Chemical space visualization and clustering with HDBSCAN and RDKit #RDKit

I caught a flu last Monday. So I stay home and rest in this week…… :-( I’m having a fever and hope will get better soon. BTW, recently I found new technique for dimension reduction called Uniform Manifold Approximation and Projection (UMAP). It was also topics in my twitter’s TL. URL links of original paper

mol encoder with Pytorch

Variable Auto Encoder (VAE) is unique method that is used for learning latent representations. VAE encodes discriminative vector to continuous vector in latent space. There are lots of examples in github. In 2016, Alán Aspuru-Guzik reported new de novo design method by using VAE. The approach represents molecules as SMLIES and SMILES strings are converted

Peptide design x Deep learning

You know recurrent neural network (RNN) is universally used in machine learning for natural language, handwriting, speech and also chemistry. Recently there are lots of reports that use RNN against SMILES strings to solve chemoinformatics problems. Today I read a short article published from Prof. Gisbert Schneider’s group. URL is below. https://pubs.acs.org/doi/10.1021/acs.jcim.7b00414 They applied RNN

Applications of Fluorine atom in Drug discovery

Some years ago, there was good review in drug discovery about the applications of fluorine. The perspective was published by researchers in BMS. There were many informations about fluorine based on their experience and published data. I think this is still useful for Med Chem. It was published in 2015 from ACS. https://pubs.acs.org/doi/abs/10.1021/acs.jmedchem.5b00258 And recently

Rational design of GPCR biased Ligand

GPCR is one of druggable target. GPCR activation controls many networks of signaling pathways, which for most receptors are mediated by both G proteins and beta-arrestins. Different signaling pathways give different effects. To avoid side effects from G protein signals, designing beta-arrestins selective ligand is useful strategy for drug discovery. And there are lots of

Think about HTS

Recently there are many approaches for starting drug discovery, i.e. FBDD, SBDD, LBDD, phenotypic, HTS and etc. High Throughput Screening (HTS) was common strategy for drug discovery when I joined my company.　That’s right even now but it is not always appropriate strategy. For pharma, it requires a lot of investment to maintain their own large

Get 3D feature descriptors from PDB file

If reader is interested in drug discovery, chemoinformatics, deep learning or MD, I think the reader might read the article below. http://pubs.acs.org/doi/abs/10.1021/acs.jcim.7b00650 KDEEP is predictor that uses Deep learning(CNN) for affinity prediction. Regarding the article, I found the new python library named HTMD ( High Through Put Molecular Dynamics ). Really I am not good