Future
In the field of biology, the application of AI has become a prominent area of interest
LinearDesign algorithm developed by Baidu It reduce the development time of mRNA vaccines
Problem
DATA is the foremost limiting factor for the application of AI in synthetic biology.
Data acquisition entails substantial investment of manpower, time, financial resources.
Solution
In order to bridge the daa gap,which represents the final frontier of AI in the field of biology.
NJU-China has innovatively employed the paradigm of
Transefer Learning: Pre-training + Fine-tuning
we focused on
the promoter sequences of eukaryotic expression systems
and train our model utilizing
a minimal amount of experimental data
To unconvered protential underlying mechanisms
Implementation
Address a critical issue in systhetic biology: Expression
LTB: an important oral vaccine adjuvant, has beem limited in its application and widespread adoption due to low expression levels.By optimizing the promoter sesuences to enhance its expression, we provide a cost-effective solution to the popularization of mucosal vaccine.
Pymaker further validates the practicality and scalability of the proposed learining paradigm and AI models in the field of synthetic biology
Higher Performance, Less Cost
Our model has much higher efficiency in
learning the working laws.
To achieve the same goodness of
fit (0.85 PCC):
requires1/10 data,
670,000 less sample size,
saves over $335,000,000
Small Sample Size Makes A Big Deal
Much stronger capability of uncover
promoter's inherent woking Laws.
with 3,000 samples:
other models only achieve a fit of 0.217,
which totally unusable.
while our model has an extraodinary fiting effect
of 0.701 of PCC,
proved to be a pure invention never been proposed.
Sponsor&Collaborator
Promotion Video
See More Info Above
