Your browser doesn't support javascript.
loading
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.
Afiliação
  • Valeri JA; Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, 02115, USA.
  • Collins KM; Institute for Medical Engineering and Science and Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA.
  • Ramesh P; Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, 02115, USA.
  • Alcantar MA; Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA.
  • Lepe BA; Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, 02115, USA.
  • Lu TK; Institute for Medical Engineering and Science and Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA.
  • Camacho DM; Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, 02115, USA.
Nat Commun ; 11(1): 5058, 2020 10 07.
Article em En | MEDLINE | ID: mdl-33028819
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.
Assuntos

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Biotecnologia / Engenharia Genética / Riboswitch / Biologia Sintética / Aprendizado Profundo Limite: Humans Idioma: En Ano de publicação: 2020 Tipo de documento: Article

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Biotecnologia / Engenharia Genética / Riboswitch / Biologia Sintética / Aprendizado Profundo Limite: Humans Idioma: En Ano de publicação: 2020 Tipo de documento: Article