Skip to main content

Questions tagged [machine-learning]

for questions about applications of machine learning algorithms to chemistry, not the machine learning methods themselves. Often bridging cheminformatics and computational chemistry, these methods consider how to represent chemical data to ML methods, accuracy thereof, and applications of chemical interest.

Filter by
Sorted by
Tagged with
0 votes
0 answers
29 views

Butina cluster over a 300 million SMILES [migrated]

I have a dataset with approximately 300 million SMILES strings, for which I want to apply Butina clustering and extract the Tanimoto similarity matrix to visualize the PCA of this matrix. Since I have ...
Manuel Ravasqueira's user avatar
5 votes
2 answers
100 views

How to get continuous spectral data into intensities of channels centered around wavelengths?

Currently, I am looking into using a spectral sensor from SparkFun in order to obtain spectroscopic data to feed into a trained neural network, however this sensor gives intensities based on inputs ...
soravoid's user avatar
1 vote
1 answer
265 views

QM9 atomization energies

I am doing experiments with the QM9 dataset from http://quantum-machine.org/datasets/ and began with some simple checks on results from the literature. Consistent with what the authors of the paper ...
Arnold Neumaier's user avatar
2 votes
1 answer
67 views

Encoding molecule bond information in matrix form

I have the following problem: I have a pentacene molecule and would like to encode its bond information into a matrix form, to feed it later on to a Machine Learning (ML) algorithm (a Graph Neural ...
ixaixim's user avatar
  • 61
1 vote
0 answers
176 views

SMILES vs. Graph for generating molecules

I have seen this question, but mine is slightly different, focusing on generating molecules. There are many published methods for generating molecules; some of which generate SMILES and others ...
Hossein's user avatar
  • 111
0 votes
0 answers
44 views

Machine learning-based drug design: how to determine the desired properties given a target?

I am working on a machine learning-based drug design project. The machine learning part was easily understood for me being from a computer science background. However, the biochemistry part is what I ...
mac179's user avatar
  • 319
3 votes
1 answer
255 views

What are the physicochemical properties related to medical drugs in the PubChem database?

I am not a chemist, in fact, I come from a computer science background. However, I am involved in a project related to artificial intelligence-based drug discovery. For this, I am trying to make a ...
mac179's user avatar
  • 319
1 vote
0 answers
109 views

Where to find the ChEMBL21 database in CSV format?

I come from an artificial intelligence (AI) background (and not from chemistry). However, my Ph.D. thesis is about Deep Reinforcement Learning for drug design. From one of the most famous papers in ...
mac179's user avatar
  • 319
5 votes
0 answers
240 views

Who named the QM7 and QM9 datasets?

On the surface, this looks like a silly question, but I honestly can't find an answer for it. Here's what I've found: The dataset was introduced in Ramakrishnan R., Dral P. et al., “Quantum chemistry ...
Imp54's user avatar
  • 51
1 vote
0 answers
125 views

Are molecular properties robust towards manipulations? [closed]

Machine learning algorithms analyze molecules by looking at their molecular representations (e. g. SMILES and graph) and turning them into feature maps that help the algorithm to distinguish them. In ...
Blade's user avatar
  • 1,503
2 votes
0 answers
200 views

The Use of AI and Machine Learning in Organic Chemistry

I am trying to do a bit of research into the current use of machine learning in chemical industry. I've been told my original question was too general so I'll try to be more specific about what I'm ...
Allod's user avatar
  • 225
2 votes
0 answers
39 views

Any large databases of IR Spectra that you are able to download many spectra in bulk [closed]

I am currently working on a machine learning project, in which I need access to many IR spectra, the more the better, in order to build a classifier that will used some unsupervised learning in order ...
finlay morrison's user avatar
3 votes
1 answer
106 views

Probabilistic molecular graph models

Let a collection of molecular graphs have at most $N$ nodes, $d$ node types (atom type), and $t$ edge types (bond types). A graph from this collection is normally represented by the tuple $(F, E)$ ...
Blade's user avatar
  • 1,503
4 votes
0 answers
52 views

Validity of graph-based molecular representation in ML

I was reading about molecular graph based generative models, which attempt to generate new molecules by training on a subset of a molecular dataset, such as QM9. In these works, there are 2 types of ...
Blade's user avatar
  • 1,503
2 votes
2 answers
153 views

What are the GDB-13 criteria for "synthetically accessible organic molecules"?

I'm trying to understand the QM datasets (QM7, QM8, QM9), however, in the description of QM7 the data is described as follows: [The] QM7 dataset, which is a subset of GDB-13 (a database of nearly 1 ...
Darien Schettler's user avatar
2 votes
1 answer
136 views

Is there anything we lose by using Google Colab for DeepChem? [closed]

DeepChem does not run on windows, so I am considering to try Google Colab. Is there any key feature I would be missing by using Google Colab remotely from Windows rather than the original DeepChem on ...
ABDULMUJEEB's user avatar
8 votes
2 answers
1k views

Is there any software that can do geometry optimization using machine learning?

DFT is a computational tool that is used in optimizing and calculating the electronic structure properties of molecules. Are there any machine learning codes that can do something similar in a ...
ABDULMUJEEB's user avatar
7 votes
2 answers
3k views

How to use Google’s Alpha-fold to predict the structure of a two-protein complex?

Alpha-fold won the CASP13 and CASP14 competitions last year and this year. It used deep learning to predict the secondary structure of a protein given the primary amino acid sequence. Google has ...
0x90's user avatar
  • 924
12 votes
2 answers
3k views

SMILES vs. Graph representation in deep learning

I have been reading papers on machine learning and deep learning methods for learning molecular space and generating molecules. These methods use different representations of the molecules. The most ...
Blade's user avatar
  • 1,503
4 votes
0 answers
82 views

Calculating the energy of a molecule using probability distributions

I wish to calculate the energy for a protein where the position of all the C-alpha atoms is known. One way is to calculate the pairwise distances between the atoms and then look up a probability ...
Chris's user avatar
  • 73
7 votes
1 answer
224 views

Is electron density a good measure of similarity between molecules?

I am looking for a good similarity measure between molecules in order to use it with machine learning algorithms. I found a paper, Carbó, R., Leyda, L. and Arnau, M. (1980), How similar is a molecule ...
Vladislav Gladkikh's user avatar
8 votes
1 answer
831 views

On diagonal terms in the Coulomb matrix

I have seen many machine learning algorithms for prediction of quantum chemistry properties that use Coulomb matrix as their input. Coulomb matrix is defined as, $$\boldsymbol{M}_{i j}^{\mathrm{...
Blade's user avatar
  • 1,503
7 votes
1 answer
321 views

Accuracy of Quantum Chemistry ML models

I am trying to compare the performance of few Quantum Chemistry property prediction ML models. I was looking at the following table from DOI: 10.1039/c7sc02664a The problem is that it does not ...
Blade's user avatar
  • 1,503
3 votes
1 answer
356 views

Molecular orbital energies prediction with ML algorithms

In the recent years, computational chemistry community has focused on Machine Learning algorithms to predict molecular properties. Unfortunately, many of the authors of such papers are not chemists, ...
Blade's user avatar
  • 1,503
5 votes
1 answer
437 views

QM9 dataset with chirality

I am trying to find chiral centers of molecules in QM9 dataset. Browsing their SMILES representation, I noticed SMILES yield using datasets.get_qm9(GGNNPreprocessor(), return_smiles=True) don't ...
Blade's user avatar
  • 1,503
4 votes
0 answers
109 views

Chirality as atom feature

I have been reading some literature on molecular energy prediction using machine learning techniques and I have noticed that one of the input parameters in many of the available models is "chirality". ...
Blade's user avatar
  • 1,503
16 votes
3 answers
2k views

Are there any datasets containing molecules with more than 38 heavy atoms?

I have been testing a machine learning approach for molecular energy prediction. The current dataset that I have is QM9, which is consist of molecules with up to 9 heavy atoms. I was wondering if ...
Blade's user avatar
  • 1,503
4 votes
1 answer
138 views

Given advancements of computational power an machine learning, how is it still not possible to predict products from chemical reactions? [closed]

Given that it seems every answer to this question is that it is "impossible to predict the outcome of a chemical reaction." Is chemistry just trial and error? Given how fundamental and revolutionary a ...
Bodhi1's user avatar
  • 49
12 votes
3 answers
930 views

Why can't equations of state be replaced by machine learning models?

The Peng-Robinson equation of state, for example, has no underlying physical meaning, and is just a model that was fit to data. Computer scientists have developed much, much better models for data ...
René's user avatar
  • 444
25 votes
4 answers
7k views

Does chemistry need machine learning? [closed]

In many fields of science (e.g. biology, medicine, psychology, statistics, physics), machine learning and artificial intelligence techniques are becoming more and more popular to analyze data. Is it ...
tag's user avatar
  • 251