Peng2555

A Short Introduction | 一句话介绍

I build computational systems for AI-driven drug discovery, with a current focus on cyclic peptides, gene editing enzymes, and reproducible biological screening pipelines.
我在做面向 AI 驱动药物发现 的计算系统，当前重点聚焦 环肽药物、基因编辑酶 与 可复现生物筛选流程。

I like research that does not stop at analysis, but becomes a usable workflow for the next experiment, the next model, and the next decision.
我希望研究工作不只停留在分析结果，而是能够沉淀成真正服务下一次实验、下一轮建模、下一步决策的工作流。

Research Map | 研究地图

What I Work On AIDD Data-centric modeling, candidate prioritization, and experiment-aware workflows. Cyclic Peptides Design, screening, permeability, developability, and structure-guided optimization. Gene Editing Enzymes Compact systems, screening logic, guide activity, safety, and delivery-aware evaluation.

我在做什么 AI 药物发现 以数据为中心的建模、候选优选与面向实验的工作流设计。 环肽药物 关注设计、筛选、通透性、成药性与结构指导优化。 基因编辑酶 关注小型系统、筛选逻辑、guide 活性、安全性与递送约束。

Signature Themes | 核心主题

01 Model-Guided Discovery Using machine learning to connect sequence, structure, and function for therapeutic discovery. 用机器学习连接序列、结构与功能，服务治疗分子发现。

02 Reproducible Pipelines Turning raw biological data into auditable workflows, structured outputs, and reusable research assets. 把原始生物数据转化为可审计流程、结构化结果与可复用科研资产。

03 Decision Systems Building systems that help prioritize experiments instead of merely producing figures. 构建真正帮助实验优先级判断的系统，而不只是生成图表。

Current Buildboard | 当前构建中

Discovery Engine
|- public dataset curation
|- sequence / structure / activity modeling
|- peptide and enzyme candidate ranking
|- experiment-facing reporting
`- reproducible research infrastructure

当前研究系统
|- 公开数据整理与质量控制
|- 序列 / 结构 / 活性建模
|- 环肽与编辑酶候选优选
|- 面向实验的结果报告
`- 可复现科研基础设施

Bento View | 个人科研画像

Scientific Direction | 科研方向 I am especially interested in problems where biological complexity meets modeling constraints: cyclic peptide design, permeability and developability, compact gene editing systems, and ML-guided screening. 我尤其关注那些“生物复杂性”与“建模约束”交汇的问题，包括环肽设计、通透性与成药性、小型基因编辑系统，以及机器学习驱动的筛选策略。	Keywords | 关键词 AIDD Cyclic peptides Computational biology Gene editing enzymes Screening systems Reproducible workflows
What I Value | 我看重什么 Clarity over complexity. Reproducibility over one-off results. Decision support over decorative analysis.	Working Style | 工作方式 Build small systems, validate assumptions, keep outputs structured, and make every project easier to continue than to restart. 用小而稳的系统推进项目，先验证假设，再扩展流程；让结果结构化，让每个项目都更容易延续，而不是反复重来。

Featured Project Directions | 代表项目方向

Direction	Description
Published gene editing enzyme reanalysis	Reanalyzing published high-impact datasets to evaluate compact editing systems and screening priorities.
Cyclic peptide AIDD workflows	Building candidate ranking pipelines for cyclic peptide design, screening, and optimization.
Bio-ML toolbox	Creating reusable utilities for computational biology, benchmarking, visualization, and structured reporting.

方向	说明
已发表基因编辑酶数据再分析	对高水平公开数据进行再分析，用于评估小型编辑系统与筛选优先级。
环肽 AIDD 工作流	构建面向环肽设计、筛选与优化的候选排序流程。
生物计算工具箱	搭建可复用的计算生物学、基准评估、可视化与结构化报告工具。

Tech Palette | 技术栈

GitHub Signal | GitHub 画像

Philosophy | 研究观

Good research code should not only run once.
It should be understandable, reproducible, and useful for the next decision.

好的科研代码不应只运行一次。
它还应当可理解、可复现，并真正服务下一步决策。

Contact | 联系方式

• GitHub: @Peng2555
• Email: 1733317481@qq.com
• ORCID: <YOUR_ORCID_LINK>
• Google Scholar: <YOUR_SCHOLAR_LINK>
• Website or Lab Page: <YOUR_WEBSITE>