NVIDIA BioIR 加速蛋白质组规模结构预测
原标题:High-Throughput Structure Prediction with BioNeMo Inference Runtime
AI 摘要
NVIDIA 发布 BioNeMo Inference Runtime(BioIR),在 NVIDIA GPU 上加速生物分子结构预测模型,同时保留 PyTorch 工作流,并支持通过 Ray 在单节点多 GPU 上运行完整模型副本以提升吞吐。该运行时已用于 AlphaFold Database 的扩展,在 4,777 个蛋白质组中加速生成约 3,100 万个候选复合物结构,其中 181 万个作为高置信度预测发布。教程展示了端到端处理器流程,包括输入准备、GPU 推理以及 PDB/mmCIF 输出,并演示如何跟踪每小时结构数和资源效率。
正文节选
Biomolecular structure prediction is now often run at proteome scale, where the goal is to move an entire worklist through the pipeline efficiently. NVIDIA BioNeMo Inference Runtime (BioIR) helps accelerate supported biomolecular structure-prediction models on NVIDIA GPUs while keeping the familiar PyTorch workflow. It uses optimized kernels and, where applicable, CUDA Graphs to speed model execution. For large batches of independent inputs, Ray can run a complete model replica on each GPU in a