用于含核糖基修饰核酸及核酸–蛋白质复合物建模的 MODNAP 基准集:以 AlphaFold 3 为例
The MODNAP benchmark set for modeling ribose-based modified nucleic acid and nucleic acid-protein complexes: a case study with AlphaFold 3.
构建 MODNAP 基准集,评估修饰核酸结构预测,并揭示 AlphaFold 3 的局部定位与几何有效性问题。
新的分析方法、软件、数据库与实验方案:解决什么问题、怎么用、和现有方法比怎样。
The MODNAP benchmark set for modeling ribose-based modified nucleic acid and nucleic acid-protein complexes: a case study with AlphaFold 3.
构建 MODNAP 基准集,评估修饰核酸结构预测,并揭示 AlphaFold 3 的局部定位与几何有效性问题。
Calibration, Uncertainty, and Data Efficiency of Deep Learning Chest Radiograph Classifiers Under Cross-Dataset Distribution Shift: A Multi-Dataset Benchmark
开展胸部X线分类器跨数据集基准评测,显示源数据校准不能稳定迁移,判别能力提升不保证概率可靠。
NeoBridge: Biophysics-informed neoantigen prioritization and clonal architecture profiling in pancreatic cancer.
提出NeoBridge用于新抗原优先级排序,并在胰腺癌中分析新抗原负荷与生存的关联。
CAR T cells from code to clinic: Framing modeling approaches with current translational research goals.
提出围绕CAR T治疗转化挑战组织建模方法的框架,为模型指导的细胞治疗设计提供路线图。
MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics.
发布MassNet十亿级质谱语料库及标准化格式,并开发用于肽段从头测序的XuanjiNovo模型。
RiboBA: a bias-aware probabilistic framework for robust ORF identification across diverse ribosome profiling protocols.
提出RiboBA,通过建模实验方案偏倚,改善周期性减弱的核糖体测序数据中的非经典ORF识别。
Innovative roadmaps for sustainable antimicrobial resistance interventions.
提出整合商业模式、发现平台、基因组监测与临床试验的路线图,推动可持续耐药性干预。
EEG Pathology Detection via Clinical Concept Bottleneck Models
提出以临床脑电生物标志物为瓶颈的模型,并揭示预测准确率与概念保真度之间的权衡。
Understanding when and how knowledge graphs add value to EHR-based clinical prediction across Parkinson’s disease, multiple sclerosis, and chronic pain
比较知识图谱整合策略,发现队列子图可节省计算成本,但预测收益取决于疾病。
Concurrent de novo inference of cell types and gene regulatory networks with NetMap
提出 NetMap,联合推断细胞级基因调控网络与细胞分群,减少对预先细胞注释的依赖。
Pocket contrastive learning and conformation fusion for robust structure-conditioned 3D molecular generation under receptor conformational variation.
提出结合口袋对比学习与参考构象特征融合的3D分子生成框架,提高受体构象变化下的稳健性。
FLIGHTED: Inferring fitness landscapes from noisy high-throughput experimental data.
提出 FLIGHTED,从含噪声的高通量实验构建概率性适应度景观,改善蛋白质适应度模型性能。
Autonomous labeling of surgical resection margins using a foundation model.
提出虚拟染墨网络VIN,在全切片图像上自动定位手术切面,减少对实体染墨的依赖。