Apple researchers have developed SimpleDesign, a novel model for protein sequence and structure codesign. Unlike previous multi-stage approaches, SimpleDesign uses a single-stage, end-to-end objective that directly models both discrete sequences and continuous structures. The model employs Transformer-based backbones for modality-specific processing and global attention, achieving competitive results on co-design and generation benchmarks after training on over 2 million sequence-structure pairs. AI
IMPACT This research could accelerate drug discovery and protein engineering by improving generative models for protein design.
RANK_REASON The item is a research paper detailing a new model for protein codesign from Apple's machine learning research division. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Apple Machine Learning Research →
- AlphaFold2
- Apple Inc.
- International Conference on Learning Representations
- International Conference on Machine Learning
- Jiarui Lu
- Jiatao Gu
- Joshua M. Susskind
- Miguel Ángel Bautista
- Mila
- Navdeep Jaitly
- SimpleDesign
- Université de Montréal
- Yizhe Zhang
- Yuyang Wang
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →