BioInsectiNet - Major Update Release
Overview
This release includes significant improvements and new features to enhance the prediction and generation capabilities of BioInsectiNet, as well as added functionalities for better compound evaluation and workflow automation.
Highlights
-
Optimized Prediction and Generation Models
Improved the affinity prediction and compound generation models for higher accuracy and faster training times. -
New Generation Examples
Added diverse and realistic examples to better showcase the compound generation capabilities and facilitate testing. -
Enhanced Genetic Algorithm
Made the genetic algorithm more robust and faster by refining mutation and selection strategies, accelerating convergence toward optimal compounds. -
Docking Algorithm Implementation
Developed a docking algorithm to simulate and visualize interactions between target FASTA sequences and generated SMILES compounds, enabling structural quality assessment. -
Automated Pipeline
Created a modular and fully automated pipeline allowing flexible execution of the workflow, either step-by-step or as a complete end-to-end process.
What's Changed
- Bump tensorflow from 2.12.0 to 2.12.1 by @dependabot[bot] in #6
- Bump keras from 2.11.0 to 2.13.1 by @dependabot[bot] in #7
Full Changelog: v1.0.1...v2.0.0
