Benchmarking Study of Deep Generative Models for Inverse Polymer Design
Molecular generative models based on deep learning have increasingly gained attention for their ability in de novo polymer design. However, there remains a knowledge gap in the thorough evaluation of these models. This benchmark study explores de novo polymer design using six popular deep generative models: Variational Autoencoder (VAE), Adversarial Autoencoder (AAE), Objective-Reinforced Generative Adversarial Networks (ORGAN), Character-level Recurrent Neural Network (CharRNN), REINVENT, and GraphINVENT. Various metrics highlighted the excellent performance of CharRNN, REINVENT, and GraphINVENT, particularly when applied to the real polymer dataset, while VAE and AAE show more advantages in generating hypothetical polymers. The CharRNN, REINVENT, and GraphINVENT models were further trained on real polymers utilizing reinforcement learning methods, targeting the generation of hypothetical polymers with high glass transition temperatures. The findings of this study provide critical insights into the capabilities and limitations of each generative model, offering valuable guidance for future endeavors in polymer design and discovery..
| Medienart: |
|
|---|
Preprint| Erscheinungsjahr: |
2024 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
chemRxiv.org - (2024) vom: 18. März Zur Gesamtaufnahme - year:2024 |
|---|
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Yue, Tianle [VerfasserIn] |
|---|
| Links: |
Volltext [kostenfrei] |
|---|
| Themen: |
|---|
| doi: |
10.26434/chemrxiv-2024-gzq4r |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
XCH042951984 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | XCH042951984 | ||
| 003 | DE-627 | ||
| 005 | 20240319103101.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 240319s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.26434/chemrxiv-2024-gzq4r |2 doi | |
| 035 | |a (DE-627)XCH042951984 | ||
| 035 | |a (chemrXiv)10.26434/chemrxiv-2024-gzq4r | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Yue, Tianle |e verfasserin |0 (orcid)0009-0005-3390-1344 |4 aut | |
| 245 | 1 | 0 | |a Benchmarking Study of Deep Generative Models for Inverse Polymer Design |
| 264 | 1 | |c 2024 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Molecular generative models based on deep learning have increasingly gained attention for their ability in de novo polymer design. However, there remains a knowledge gap in the thorough evaluation of these models. This benchmark study explores de novo polymer design using six popular deep generative models: Variational Autoencoder (VAE), Adversarial Autoencoder (AAE), Objective-Reinforced Generative Adversarial Networks (ORGAN), Character-level Recurrent Neural Network (CharRNN), REINVENT, and GraphINVENT. Various metrics highlighted the excellent performance of CharRNN, REINVENT, and GraphINVENT, particularly when applied to the real polymer dataset, while VAE and AAE show more advantages in generating hypothetical polymers. The CharRNN, REINVENT, and GraphINVENT models were further trained on real polymers utilizing reinforcement learning methods, targeting the generation of hypothetical polymers with high glass transition temperatures. The findings of this study provide critical insights into the capabilities and limitations of each generative model, offering valuable guidance for future endeavors in polymer design and discovery. | ||
| 650 | 4 | |a Chemistry |7 (dpeaa)DE-84 | |
| 650 | 4 | |a 540 |7 (dpeaa)DE-84 | |
| 700 | 1 | |a Tao, Lei |4 aut | |
| 700 | 1 | |a Varshney, Vikas |4 aut | |
| 700 | 1 | |a Li, Ying |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t chemRxiv.org |g (2024) vom: 18. März |
| 773 | 1 | 8 | |g year:2024 |g day:18 |g month:03 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.26434/chemrxiv-2024-gzq4r |z kostenfrei |3 Volltext |
| 912 | |a GBV_XCH | ||
| 951 | |a AR | ||
| 952 | |j 2024 |b 18 |c 03 | ||