Generate new molecules from fragments with Diffusion model #cheminformatics #rdkit #difflinker #memo

Designing linked molecule from fragments is one of the important task for drug desing such as FBDD, Scaffold hopping (e.g. replace core) and PROTAC molecule design. As readers know there are lots of solutions to do it, for examoke BROOD is one of the famous commercial package for fragment replacement. I can’t use commercial package

Edit atom indices of RDKit Mol object #memo #cheminformatics

Atom indecies are unique number of each atom. And RDKit adds the index when mol object is generated. RDKit makes mol object from SMILES, Inchi, molblock and lots of formats. To make canonical representation of molecules, the atom indices are asigned allways same roules in automatically. The indices are asigned regardless of scaffold. So if

Current status of DMTA cycle in AZ #memo #DDT #publication #AI/ML

March is end of fiscal year in most of Japanese company. I spent lots of time for paper work in these days… ;P As many readers know that, DMTA cycle is key of drug discovery/optimization process and lots of computational and high thoughput experimental apporaches are available in the process recenlty. I think AZ is

Predict protein-ligand complex dynamics with python #dynamicbind #cheminformatics #memo

Computer aided drug design is one of the powerful approach for drug discovery these days. Docking study of target protein and ligands is common proceduer to evaluate whether the compound fit target protein’s pocket or not. However there is a limitation in the method. Most of the docking apporach handle protein and ligand as rigid

Update rdkit/shape-it #RDKit #shape-it #cheminformatics

Today I tried to build rdkit/shape-it because I could not build shape-it with current version of rdkit. So I struggled error message to fix the issue.(I’m not so good at C++ ;P) There are two issues in the current code.1. Version of c++ in the CMakeLists.txt is old, so I changed it from c++14 to

Visualize feature importance with marimo #cheminformatics #RDKit #marimo

I posted new generation of notebook, marimo recently. It is cool and easy to make interactive analysis environment with python. I’m interested in the package and am thinking how to use in chemoinformatics tasks. In QSAR tasks, chemoinformaticians are often asked the reason of prediction of the model. So XAI (explainable AI) is an attractive

New ML package for cheminformatics #cheminformatics #QSAR #ML

I introduced scikit-mol in my blog post before. The package integrates scikit-learn and rdkit. It’s easy to use because user can build QSAR model from scikit-learn’s API. I like the package. And recently I found another useful package for cheminformatics named ‘molflux‘ witch is developed by researchers in Exsicentia, famous AI Drug Discovery pharma. molflux

New type of python notebook #marimo #cheminformatics #RDKit

Jupyter-lab, Jupyter-note book, streamlt and other packages are useful for data science beucase it can analyze and visualize data step by step. I like streamlit and dash for making simple web app. And some days ago I found new and cool package named marimo. From the documentaion, marimo is an open-source reactive notebook for Python — reproducible,

Try to use mmpdb v3.1 #RDKit #MMPDB #cheminformatics

As many RDKitters know that Andrew released new version of MMPDB! Recent version of MMPDB has lots of useful methods, one of the generate method it can generate new molecules from given smiles. The method can generate not only all possible molecules from MMPDB but also constrained molecules with options, ‘–query’ and ‘–constant’. ‘–query’ option

ChEMBL multitask prediction with python requrests #memo #chembl #cheminfo

ChEMBL provides multitask prediction model from its github repo. And shared useful blog post.https://chembl.blogspot.com/2019/05/multi-task-neural-network-on-chembl.html By using the code, we can get predicted target list from given molecules. And the prediction can run python, C++, JS and Knime! ChEMBL team provides not only source code but also predicted results when we search compound in ChEMBL DB.

Useful package for filtering molecules of python #RDKit #Python #memo

I wrote blog post of how to use Lilly filter from REINVENT4. In the post, I build lilly filter from source code. After posted, Hadrien introduced me a package for using lilly filter which can call from python. Thanks! The package is provided from datamol-io’s reposidory named medchem. I had intreste the package so I installed

Run virtual screening with only python! #chemoinformatics #SBDD

If you have protein-ligand co- crystal structure at the early stage of drug discovery, SBDD based approach is useful I think. As many readers know that there are lots of useful commercial packages to conduct SBDD. But also there are useful packages at Open science area. One of the famous and useful OSS for SBDD

Regist new molecules with lwreg from web app #RDKit #lwreg #cheminformatics

Now there is RDKit based chemical cartridge for sqlite3 and posgresql. So it’s useful for developping cheminfo web app with these databases. By using these databases with web app, we need to defne the schema of databases at first. Also many cheminformaticians handle lots of molecules. So it’s useful to register their idea or handled

Visualize result of molshap #RDKit #chemoinformatics #memo

Finding the best combination of substituents is very important task for compound optimization. Recently there are lots of methods to predict the combination. By using Deep learning or other complex methods are difficult understand for chemists. So I like Free Wilson analysis, because it’s simple but easy to understand because FW analysis uses liner regression.

Update Scikitlearn and visualize chemical space with rdkit #RDKit #Chemoinformatics #Scikitlearn

As most of readers know that new version of scikitlearn is released ;) There are lots of improvements are implemented. And you can see the details in original documenation.https://scikit-learn.org/stable/whats_new/v1.3.html One of interesting news is that scikit-learn v13 has implemented HDBSCAN. The origainal article is following link. And it implemented indpendend package of python.Article: https://link.springer.com/chapter/10.1007/978-3-642-37456-2_14Package: https://github.com/scikit-learn-contrib/hdbscan

Some updates of rdkit_cli #RDKit #chemoinformatics #typer

I enjoyed a user group meeting last week I could have lots of useful discussion there thanks for all presenters and perticipants ;) BTW, I’m still enjoying to make RDKit based CLI tool and I added some routine task to the code. Of course some function is already impremented other tool such as openbabel etc.

Develop rdkit-cli tool with typer #RDKit #chemoinformatics #typer

Recently I’m learning not only coding but also no-code tool such as knime. It’s important to chatch up wide range of informatics IMHO…. BTW, as you know rdkit has lots of useful tools. And there are many useful packages which depend on RDKit. It’s because rdkit is growing very rapidly. Today I learned useful package

Make learning process with human in the loop #optuna #memo #python #rdkit

Many chemofinromaticians have checked Greg’s great blog post which describes how to draw draw molecules in various way. https://greglandrum.github.io/rdkit-blog/posts/2023-05-26-drawing-options-explained.html Rendering molecule doesn’t directly contribute drug design but it’s really importnt for us because medicinal chemists have their own preferences for drawing moleucles such as font size, highlight color etc, etc. It’s too difficult to parametarize

Optimize ML model with few line code #chemoinformatics #ML #RDKit #falcon

Machine learning is democratizing, and there are many great libraries available in Python such as scikit-lean pycaret. Recenlty we can build ML model without writing lots of code. Today I found new package named ‘falcon‘ which can build optimized model few lines. One of the famous package for AutoML is auto-scikitlearn, the package is also

Use jupyternotebook from knime #knime #chemoinformatics #memo

Recently I posted an example to build custom knime node with python. It was a nice experience for me. And I’m still reading knime documentation. Today I tried to use jupyter notebook from knime. There is a good example to do that with workflow and dataset.https://docs.knime.com/latest/python_installation_guide/index.html#jupyter-notebooks The example described avobe reads dataset and do some