Table 2 Machine learning models for antibiotic discovery.

From: Accelerating antibiotic discovery through artificial intelligence

 

Public release

 

Algorithm

Code

Data

Software

Software type

Antimicrobial activity prediction

    

 Artificial neural network40

 

Yes

  

 Support vector machine38

 

Yes

  

 Multinomial logistic regression33

 

Yes

  

 LSTM RNN44

Yes

Yes

Yes

Command-line tool

 XGBoost42

Yes

Yes

Yes

Command-line tool

 Directed-message passing neural network16

Yes

Yes

Yes

Web server, Docker container

 DBSCAN47

 

Yes

Yes

Web server

 DBSCAN48

  

Yes

Web server

 Convolutional neural network41

 

Yes

Yes

Web server

 Generalized linear model49

    

 Random forest50

    

Hemolytic activity prediction

    

 Classification and regression trees55

 

Yes

  

 Artificial neural network54

 

Yes

Yes

Web server

 Gradient boosting classifiers56

Yes

Yes

  

 Support vector machine183

 

Yes

Yes

Web server, mobile app, standalone

De novo antibiotic design

    

 Variational autoencoder45

 

Yes

  

 LSTM RNN30

Yes

Yes

Yes

Command-line tool

 LSTM RNN120

    

 Generative adversarial network119

Yes

Yes

Yes

Command-line tool

  1. Machine learning models cited in this review pertain specifically to antimicrobial compound discovery, i.e., those that predict antimicrobial activity, those trained on antimicrobial compound data to predict drug-likeness, and those that generate potential antimicrobials. Public release of model source code, training and/or testing data, and/or associated software tools are noted. Criteria for data release were lenient, with “yes” indicating partial or full release of training or testing data.