-
Oct 29, 2024 | 
nature.com | Bin Shao |Jiawei Yan
AbstractInspired by the success of large language models (LLMs), we develop a long-context generative model for genomes. Our multiscale transformer model, megaDNA, is pre-trained on unannotated bacteriophage genomes with nucleotide-level tokenization. We demonstrate the foundational capabilities of our model including the prediction of essential genes, genetic variant effects, regulatory element activity and taxonomy of unannotated sequences.
-
Oct 1, 2024 | 
biorxiv.org | Bin Shao
AbstractWe introduce PlasmidGPT, a generative language model pretrained on 153k engineered plasmid sequences from Addgene. PlasmidGPT generates de novo sequences that share similar characteristics with engineered plasmids but show low sequence identity to the training data. We demonstrate its ability to generate plasmids in a controlled manner based on the input sequence or specific design constraint.
-
May 24, 2024 | 
onlinelibrary.wiley.com | Bin Shao |Jiabin Liu |Heng C. Su |Yingying Zong
7075 aluminum alloy is widely used in the aerospace field because of its low density, high specific strength, high fracture toughness, and good machinability. This paper systematically studied the evolution of the microstructure and properties of 7075 aluminum alloy at different temperatures. Based on the rolling and heat treatment process, a short process of rolling and heat treatment is proposed, which can significantly improve the properties.
-
Mar 29, 2024 | 
link.aps.org | Haiping Li |Jian Zou |Bin Shao
We consider a collision model representation of nonequilibrium dynamics for an externally driven open quantum system. Specifically we investigate the nondemolition quasiprobability distributions (QPDFs) of work and heat in both Markovian and non-Markovian regimes. In this model, work and heat correspond to different evolution processes, and their contributions can be distinguished.
-
Mar 4, 2024 | 
nature.com | Bin Shao
AbstractTranslation elongation is essential for maintaining cellular proteostasis, and alterations in the translational landscape are associated with a range of diseases. Ribosome profiling allows detailed measurements of translation at the genome scale. However, it remains unclear how to disentangle biological variations from technical artifacts in these data and identify sequence determinants of translation dysregulation.
-
Feb 29, 2024 | 
microsoft.com | Tong Wang |Bin Shao |Tie-Yan Liu |Alyssa Hughes
Molecular geometry modeling is a powerful tool for understanding the intricate relationships between molecular structure and biological activity – a field known as structure-activity relationships (SAR). The main premise of SAR is that the biological activity of a molecule is dictated by its specific chemical structure, not only the connections between nuclei but also how the molecule is twisted and arranged in a three-dimensional configuration.
-
Feb 7, 2024 | 
biorxiv.org | Bin Shao
AbstractInspired by the success of large language models, we develop a long-context generative model for genomes. Our multiscale transformer model was pre-trained on unannotated bacteriophage genomes with byte-level tokenization. We demonstrate the foundational capabilities of our model including the prediction of essential genes, genetic variant effects, regulatory element activity and taxonomy of unannotated sequences.
-
Jan 14, 2024 | 
biorxiv.org | Bin Shao
AbstractGenerative pre-trained transformers (GPTs) have revolutionized the field of natural language processing. Inspired by this success, we develop a long-context generative model for genomes. Our multiscale transformer model was pre-trained on unannotated bacteriophage genomes with byte-level tokenization. It generates de novo sequences up to 96K base pair with functional genomic structure, including regulatory elements and novel proteins with phage-related functions.
-
Dec 27, 2023 | 
onlinelibrary.wiley.com | Heng C. Su |Bin Shao |Jiabin Liu |Yingying Zong
Wire arc additive manufacturing technology of 7050 aluminum alloy has a broad application prospect in aerospace field. But the lack of comprehensive properties caused by coarse columnar grains and pores has become an important bottleneck. In this study, hot compression was conducted to simulate forging process to investigate the microstructure improvement of as-deposited 7050 Al alloy by plastic deformation.
-
Dec 19, 2023 | 
biorxiv.org | Bin Shao
AbstractGenerative pre-trained transformers (GPTs) have revolutionized the field of natural language processing. Inspired by this success, we develop a long-context generative model for genomes. Our multiscale transformer model was pre-trained on unannotated bacteriophage genomes with byte-level tokenization. It generates de novo sequences up to 96K with functional genomic structure, including regulatory elements and novel proteins with phage-related functions.
