Automatic chemical design using a data-driven continuous representation of molecules · HackerPapers discussions

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by dekhn

Wed, 05 Nov 2025 23:31:54 GMT

Let me start with an example- some time ago I worked on a VAE that encoded and decoded SMILES strings. The idea is that you should be able to encode a SMILES into an embedding space, do all the normal things you would do in that space, and then convert the resulting embedding vector back to a valid molecule. The VAE is trained with a very large number of valid SMILES strings, typically tokenized at the character level (so "C" is a token, and "Br" is "B" then "r"). I and others have observed that VAEs trained like this produce large number of embedding vectors that do not decode to valid…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by dekhn

Mon, 01 Jul 2024 02:18:02 GMT

No, I mean the older paper with the VAE for SMILES https://arxiv.org/abs/1610.02415 which we ended up implementing here: https://github.com/maxhodak/keras-molecules SELFIES is just a pip install away; I think the team learned from their previous papers to be more open about the limits, as well as releasing code and weights for reproducibility. I don't follow this area closely enough to really have any comments about SELFIES other than "well, at least the problems we identified were addressed in later work". Specifically, my goal was to start with two SMILES strings, encode them to vectors…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by gpcr1949

Fri, 25 Jun 2021 21:24:55 GMT

It's important to note that although chemical space is quite large, most of this space is not easy to synthesize and also is not chemically feasible, stable or desirable. Another interesting "small" subset of chemical space is ZINC [0] which is a database of about a billion commercially offered compounds, meaning that manufacturers at a minimum think they can easily make them (and effectively the fulfilment is quite high when random compounds are ordered, e.g. 95% in this paper where they did molecular docking simulations on the entirity of this database to find new melatonin receptor…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by 41321121

Sun, 22 Jan 2017 04:31:42 GMT

It seems to me a fundamental problem with this paper is that they're training an autoencoder on a relatively small set of drugs that have been tested on cancer. They are then trying to approximately index a much larger dataset of chemicals and claiming these are novel possible cancer treatments because some of them have been previously considered as cancer drugs. It's not clear that this is doing much more than finding drugs similar to the training drugs in the new dataset. Given that a large part of pharmaceutical/chemical development is based on slightly modifying existing compounds, it is…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by momeara

Tue, 25 Oct 2016 13:32:20 GMT

Several people have asked for background material for the workshop-- (Wallach, 2015, http://arxiv.org/pdf/1510.02855.pdf) AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery (Duvenaud, 2015, http://papers.nips.cc/paper/5954-convolutional-networks-on-g...) Convolutional Networks on Graphs for Learning Molecular Fingerprints (Kearnes, 2016, http://arxiv.org/abs/1606.08793v1) Modeling Industrial ADMET Data with Multitask Networks (Gómez-Bombarelli, 2016, doi:10.1038/nmat4717) Design of efficient molecular organic light-emitting diodes by a…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by duvenaud

Tue, 25 Oct 2016 03:24:42 GMT

One of the authors of the original paper https://arxiv.org/abs/1610.02415 here. From a machine learning point of view, we simply glued together two techniques: text autoencoders and Bayesian optimization. That is, we trained an autoencoder to transform a text representation of chemicals (SMILES) to and from continuous vectors. Then “chemical design” is just maximizing a function of a continuous variable, something that we already know a lot about. We also showed off some of the nice things that one can do with continuous latent representations, such as interpolation. This had already been…

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — mentioned by cing

Sun, 16 Oct 2016 03:45:49 GMT

People are working on such things! https://arxiv.org/abs/1610.02415

Paper and discussion history. A mention is not an endorsement.

Automatic chemical design using a data-driven continuous representation of molecules — HN submission

Thu, 13 Oct 2016 06:42:00 GMT

Automatic chemical design using deep neural networks

4 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.