免疫原性
计算机科学
计算生物学
管道(软件)
注释
MHC I级
表位
稳健性(进化)
加权
主要组织相容性复合体
人类白细胞抗原
复制
肽
数据挖掘
抗菌剂
杠杆(统计)
人工智能
班级(哲学)
生物
作者
Rajasekaran R,Akshay Sasidharan,Shreya Nandi Roy,Ananya Pandey,V. R. Sujatha,Kanishk Prabhu
标识
DOI:10.5281/zenodo.22653865
摘要
This repository contains a set of Python scripts implementing a multi-stage computational pipeline for screening antimicrobial peptide (AMP) candidates for safety and immunogenicity risk. The pipeline was developed and applied to fish-derived AMPs retrieved from the Antimicrobial Peptide Database (APD6) and is described in full in the associated publication. The pipeline performs: (1) sequential hard-exclusion safety filtering for toxicity (ToxinPred), hemolytic activity (HemoPI), and allergenicity (AlgPred 2.0) (2) IL-10 induction annotation (3) extraction of MHC class II and MHC class I binding metrics from NetMHCIIpan-4.0 and NetMHCpan-4.1 prediction outputs (4) linear B-cell epitope coverage extraction from BepiPred 2.0 reports (5) weighted composite safety/immunogenicity risk scoring with MHC-II-conditioned B-cell risk; and (6) one-at-a-time and Monte Carlo sensitivity analyses to assess robustness of the composite weighting scheme. Although developed for fish-derived AMPs, the scripts are written as general templates and can be adapted to any AMP candidate list with equivalent prediction-tool outputs. All modelling choices (weights, thresholds, conditioning parameters) are documented in each script's docstring and flagged for reconsideration before reuse.
科研通智能强力驱动
Strongly Powered by AbleSci AI