嵌入向量相似性1 − 余弦距离 / 2 — 始终可用。
NW 一致性Needleman-Wunsch 全局序列一致性 [0, 1]。
SW 一致性Smith-Waterman 局部序列一致性 [0, 1]。
证据权重参考注释的 GO 证据代码质量。
分类邻近度1 / (1 + 分类距离)
linear: Σ(w_i · s_i) / Σ(w_i) over all active (w_i > 0 and signal available) signals.
evidence_weighted: Same as linear, then multiplied by the resolved evidence weight — down-ranks IEA even when other signals are strong.
