Sequence-to-function deep learning frameworks for engineered riboregulators
© 2020, The Author(s). While synthetic biology has revolutionized our approaches to medicine, agriculture, and energy, the design of completely novel biological circuit components beyond naturally-derived templates remains challenging due to poorly understood design rules. Toehold switches, which ar...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer Science and Business Media LLC
2021
|
| Online Access: | https://hdl.handle.net/1721.1/134432 |
| _version_ | 1811078935740416000 |
|---|---|
| author | Valeri, Jacqueline A Collins, Katherine M Ramesh, Pradeep Alcantar, Miguel A Lepe, Bianca A Lu, Timothy K Camacho, Diogo M |
| author2 | Massachusetts Institute of Technology. Institute for Medical Engineering & Science |
| author_facet | Massachusetts Institute of Technology. Institute for Medical Engineering & Science Valeri, Jacqueline A Collins, Katherine M Ramesh, Pradeep Alcantar, Miguel A Lepe, Bianca A Lu, Timothy K Camacho, Diogo M |
| author_sort | Valeri, Jacqueline A |
| collection | MIT |
| description | © 2020, The Author(s). While synthetic biology has revolutionized our approaches to medicine, agriculture, and energy, the design of completely novel biological circuit components beyond naturally-derived templates remains challenging due to poorly understood design rules. Toehold switches, which are programmable nucleic acid sensors, face an analogous design bottleneck; our limited understanding of how sequence impacts functionality often necessitates expensive, time-consuming screens to identify effective switches. Here, we introduce Sequence-based Toehold Optimization and Redesign Model (STORM) and Nucleic-Acid Speech (NuSpeak), two orthogonal and synergistic deep learning architectures to characterize and optimize toeholds. Applying techniques from computer vision and natural language processing, we ‘un-box’ our models using convolutional filters, attention maps, and in silico mutagenesis. Through transfer-learning, we redesign sub-optimal toehold sensors, even with sparse training data, experimentally validating their improved performance. This work provides sequence-to-function deep learning frameworks for toehold selection and design, augmenting our ability to construct potent biological circuit components and precision diagnostics. |
| first_indexed | 2024-09-23T11:07:34Z |
| format | Article |
| id | mit-1721.1/134432 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:07:34Z |
| publishDate | 2021 |
| publisher | Springer Science and Business Media LLC |
| record_format | dspace |
| spelling | mit-1721.1/1344322024-03-19T17:34:56Z Sequence-to-function deep learning frameworks for engineered riboregulators Valeri, Jacqueline A Collins, Katherine M Ramesh, Pradeep Alcantar, Miguel A Lepe, Bianca A Lu, Timothy K Camacho, Diogo M Massachusetts Institute of Technology. Institute for Medical Engineering & Science Massachusetts Institute of Technology. Department of Biological Engineering Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Massachusetts Institute of Technology. Research Laboratory of Electronics © 2020, The Author(s). While synthetic biology has revolutionized our approaches to medicine, agriculture, and energy, the design of completely novel biological circuit components beyond naturally-derived templates remains challenging due to poorly understood design rules. Toehold switches, which are programmable nucleic acid sensors, face an analogous design bottleneck; our limited understanding of how sequence impacts functionality often necessitates expensive, time-consuming screens to identify effective switches. Here, we introduce Sequence-based Toehold Optimization and Redesign Model (STORM) and Nucleic-Acid Speech (NuSpeak), two orthogonal and synergistic deep learning architectures to characterize and optimize toeholds. Applying techniques from computer vision and natural language processing, we ‘un-box’ our models using convolutional filters, attention maps, and in silico mutagenesis. Through transfer-learning, we redesign sub-optimal toehold sensors, even with sparse training data, experimentally validating their improved performance. This work provides sequence-to-function deep learning frameworks for toehold selection and design, augmenting our ability to construct potent biological circuit components and precision diagnostics. 2021-10-27T20:04:59Z 2021-10-27T20:04:59Z 2020 2021-01-28T19:43:16Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/134432 en 10.1038/s41467-020-18676-2 Nature Communications Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/ application/pdf Springer Science and Business Media LLC Nature |
| spellingShingle | Valeri, Jacqueline A Collins, Katherine M Ramesh, Pradeep Alcantar, Miguel A Lepe, Bianca A Lu, Timothy K Camacho, Diogo M Sequence-to-function deep learning frameworks for engineered riboregulators |
| title | Sequence-to-function deep learning frameworks for engineered riboregulators |
| title_full | Sequence-to-function deep learning frameworks for engineered riboregulators |
| title_fullStr | Sequence-to-function deep learning frameworks for engineered riboregulators |
| title_full_unstemmed | Sequence-to-function deep learning frameworks for engineered riboregulators |
| title_short | Sequence-to-function deep learning frameworks for engineered riboregulators |
| title_sort | sequence to function deep learning frameworks for engineered riboregulators |
| url | https://hdl.handle.net/1721.1/134432 |
| work_keys_str_mv | AT valerijacquelinea sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT collinskatherinem sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT rameshpradeep sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT alcantarmiguela sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT lepebiancaa sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT lutimothyk sequencetofunctiondeeplearningframeworksforengineeredriboregulators AT camachodiogom sequencetofunctiondeeplearningframeworksforengineeredriboregulators |