> For the complete documentation index, see [llms.txt](https://docs.hits.ai/hyperlab-release-note-en/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hits.ai/hyperlab-release-note-en/changelog_en/2025-04-10-release-note.md).
# 2025-04-10 Release Note
##  Changelogs
### 1. New Feature: Hyper Binding Co-folding
**Hyper Binding Co-folding** is a new feature that supports everything from sequence-based protein registration to protein-ligand interaction analysis using AI-powered structure prediction technology.\
Inspired by AlphaFold-style predictions, this feature is designed to be researcher-friendly, enhancing both precision and flexibility in experimental design and drug discovery.
**Key Features of Co-folding**:
#### **(1) Sequence-based Protein Registration**
You can register target proteins simply by entering their name. Even in the absence of PDB information, 3D structures can be predicted using only the sequence.\
AI-powered prediction enables the analysis of proteins that have not yet been experimentally resolved, making it highly useful in early-stage research.
For more information, please refer to the **Protein Management** section in the HyperLab Docs.
{% embed url="" %}
#### **(2) Structural Changes After Drug Binding**
Using the 3D Viewer, you can analyze structural changes in the protein-ligand complex after binding, allowing for deeper and more biologically relevant insights beyond static structure analysis.
For more information, refer to the **3D Viewer** section in the HyperLab Docs.
{% embed url="" %}
### 2. New Feature: Hyper Screening X
A new feature, **Hyper Screening X**, has been added to the left menu of LabSpace.
**Hyper Screening X** generates molecules with high synthetic accessibility and binding affinity in an ultra-large virtual chemical space built from commercial building blocks and common reactions.\
You can also request synthesis directly for molecules generated through Hyper Screening X.
**Key Features of Hyper Screening X**:
#### **(1) Model Management**
Train and manage new AI models to generate molecules using Hyper Screening X.\
Protein structure is required for model training. You can also define the desired range of physicochemical properties for the molecules to guide model optimization.
#### **(2) Molecule Generation**
Use trained AI models to generate molecules, track the generation process, and manage the results.\
You can run multiple generations using a single model and even review the synthesis pathways of the generated molecules.
#### **(3) Cart**
Add generated molecules to a shopping cart and request synthesis quotations directly.
For more information, refer to the **HyperLab User Guide: Hyper Screening X**.
{% embed url=".." %}
### 3. New Feature: AI Assistant
We’ve added a new **AI assistant** feature to help you use HyperLab more easily and effectively for your research goals.
With HyperLab’s AI Assistant, you can conduct research more efficiently and sustainably.
For more information, refer to the **HyperLab User Guide: AI assistant.**
{% embed url="" %}
### 4. Hyper Binding Upgrade
For proteins registered as **3D Structure** type, the box size you set is now more precisely reflected in calculations.
> ⚠️ **Note**: Calculation results may differ even under the same conditions as before.
***
##  Bug fix
Bug fixes and stability improvements have been made.
