End-to-end solutions

Phase 1: Data-first, AI-native research and development

Why can't data-driven studies be more accessible?

Biology is so complicated that its R&D can largely benefit from data intensive approach. It requires high-throughput screening, deep learning models, AI-aided design, etc.

None of those platforms is easy or cheap to build. Organizing experts in AI and biology is not easy either, both technology and culture wise.

In phase 1, we provide data-driven solutions to streamline the R&D processes for protein engineering. It integrates AI algorithms, ultra-high-throughput screening, large part registry and pre-trained models. So you could focus on proteins themselves.

Turnaround time: Starting from 6 months.
Deliverables: Technical reports with raw datasets

Phase 2: Flexible protein production

Why can't protein production be easier?

After finishing the R&D, you may want to produce your proteins to test with the downstream applications. Although recombinant protein production seems simple in principle, it was never easy in practice:

50% proteins failed to express (Bhandari et al. 2021)
20% proteins failed to purify (Braun et al. 2001)

In addition, scaling up bioproduction is technically difficult. And it is even worse when adapting the production scale to demands.

In phase 2, we provide solutions to produce and purify desired proteins on a flexible scale. So you don't need to worry about building the bioreactors, optimizing the processing, or managing the production scales.

Turnaround time: Starting from 3 month.
Deliverables: Protein (>80% purity) at target amount with QC reports

Technological highlights

Protein Design with Large AIs

We use the state-of-the-art large deep learning models to develop AI algorithms for protein designs:

Based on parallel protein structure folding
Based on large language models (LLM)
Based on pretrained seq-to-function models

>> Example: In silico evolution with Ailurus Fold™

Protein and Strain Design with Large Part Registry

We prebuilt and maintain a large collection of biological components that can be reused for combinatorial screening.

>> Example: Ailurus vec®️ - 1.2 trillion pre-built vectors
>> Example: 20,000+ pre-built protein building blocks

Ultra-High-Throughput Screening Assays in Parallel

We design cells as micro labs that perform measurement assays, which enables ultra-high-throughput screening in parallel at low costs.

>> Example: recording expression levels into DNA storages

Protein Production and Purification at scale

We design cells as micro reactors with self-control of fermentation and downstream processing, which enable high-throughput protein production and purification in parallel at scales.

>> Example: PandaPure®️ - autonomous protein purification

Make a request

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.