用大模型推理筛选基因，提升癌症生物标志物发现的准确性。
Mamba-SSM with LLM Reasoning for Feature Selection: Faithfulness-Aware Biomarker Discovery
- 用Mamba SSM提取基因重要性，再由大模型逐条分析过滤混杂因素。
- 最终17个基因的预测性能超越5000个基因基线，AUC达0.927。
- 适合关注精准生物标志物筛选与可解释性的临床研究者。
深度序列模型的梯度显著性可快速识别候选生物标志物，但结果常受组织成分混杂因素干扰，影响下游分类器性能。本文研究大语言模型链式思维（CoT）推理是否能有效过滤此类混杂因素，以及推理质量与下游性能的关系。在TCGA-BRCA RNA-seq数据上训练Mamba SSM，通过梯度显著性提取前50个基因；随后使用DeepSeek-R1对每个候选基因进行结构化链式思考，生成最终17个基因集。在独立测试集上，原始50基因显著性集合表现劣于5000基因方差基线（AUC 0.832 vs. 0.903），而经大模型过滤后的17基因集则优于基线（AUC 0.927），仅使用294倍更少特征。信仰度审计（COSMIC CGC、OncoKB、PAM50）显示，17个基因中有6个（35.3%）为已验证的乳腺癌生物标志物，但输入中16个已知乳腺癌基因有10个被遗漏，包括FOXA1。该表现与推理忠实度之间的差异表明，在此场景下存在选择性忠实：针对性去除混杂因素可提升预测性能，无需全面召回所有已知基因。
原文摘要 · Abstract (English)
Gradient saliency from deep sequence models surfaces candidate biomarkers efficiently, but the resulting gene lists can be contaminated by tissue-composition confounders that degrade downstream classifiers. We study whether LLM chain-of-thought (CoT) reasoning can filter these confounders, and whether reasoning quality is associated with downstream performance. We train a Mamba SSM on TCGA-BRCA RNA-seq and extract the top-50 genes by gradient saliency; DeepSeek-R1 evaluates every candidate with structured CoT to produce a final 17-gene set. On the held-out test split, the raw 50-gene saliency set (no LLM) performs worse than a 5,000-gene variance baseline (AUC 0.832 vs. 0.903), while the LLM-filtered set surpasses it (AUC 0.927), using 294x fewer features. A faithfulness audit (COSMIC CGC, OncoKB, PAM50) shows that 6 of 17 selected genes (35.3%) are validated BRCA biomarkers, while 10 of 16 known BRCA genes present in the input were missed - including FOXA1. This divergence between downstream performance and reasoning faithfulness suggests selective faithfulness in this setting: targeted confounder removal can improve predictive performance without comprehensive recall.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。