.png){kind=icon}
.png){kind=icon}
Return
SegDesign: A modular framework for controllable protein segment engineering
.png){kind=icon}
OA
AI
DOI：10.1002/pro.70542.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Local protein structure engineering, such as targeted remodeling of loops or flexible regions, is critical for mechanistic studies and protein optimization but remains challenging to perform in a controllable and reproducible manner. Despite advances in protein foundation models and generative structure design, most existing methods emphasize global scaffold generation and offer limited support for precise, region-specific intervention while preserving the overall fold. Here we present SegDesign, a modular framework for segment-level protein engineering that integrates backbone reconstruction, sequence redesign, and multi-stage structural evaluation into a unified workflow. SegDesign enables user-defined secondary-structure manipulation of selected regions and produces traceable, experimentally testable variant panels. Applied to Gallus gallus terminal deoxynucleotidyl transferase (GgTdT), SegDesign converts an intrinsically flexible loop into either a stable alpha-helix or beta-strand without disrupting the global fold, as supported by AlphaFold3 modeling. Applications to six additional enzymes further demonstrate the generality of the approach. SegDesign establishes an accessible paradigm for controllable local protein structure engineering in both mechanistic and applied settings. The standalone implementation of SegDesign is publicly available at https://github.com/mike114b/Segdesign.
Keywords:
local structural remodeling
protein backbone remodeling
protein design framework
protein engineering workflows
protein segment engineering
secondary-structure control
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.2
Papers:
732
Citations:
2.2W