1. Capsid Sequence Library

Hundreds of thousands of capsid sequences are designed, synthesized on a DNA printer, and assembled into a capsid library to prepare for screening. This allows the preparation of a training data set.

2. NGS & Preditors

Among the libraries produced, AAVs with multiple characteristics important for treatment success are selected through next-generation sequencing. A process of selecting suitable AAV candidates is carried out based on analysis and prediction.

3. Machine Learning Algorithms

A machine learning model is trained to predict the function of new base sequences, and a map suitable for AAV capsid selection is built. A more sophisticated and detailed database can be obtained through repeated learning.

4. In vivo assay

Among the AAV candidates selected as suitable, the produced capsids undergo cross-validation model assessment selection training through in vivo assay.

5. Hit NEW AAV Serotype

Through the above processes, you can obtain the most suitable candidate AAV that meets the customer's request based on the efficiency, specificity, immune response evasion, and manufacturability of in vivo delivery.

1. Promoter enginerring

Develop cell/tissue-specific promoters through machine learning-based screening.

2. Gene auto regulation

Resolve the safety and sustainability of therapeutics through technology that can appropriately control the intracellular expression of gene therapy.

1. AAV vector design

Maximize specificity and expression efficiency for target tissues through AAV capsid selection, promoter optimization, and transgene size control.

2. Screening and optimization of gene expression conditions

Test the expression of AAV vectors in various cell lines and culture conditions to determine optimal production conditions. Maximize AAV production by adjusting parameters such as media composition, temperature, and pH.

3. In vitro/in vivo test

4. Validation of effective substances

5. Drug optimization