He Adam Y, Palamuttam Nathan P, Danko Charles G
Cornell University, Ithaca, NY 14850.
bioRxiv. 2025 Feb 15:2024.10.15.618510. doi: 10.1101/2024.10.15.618510.
Sequence-to-function models have broad applications in interpreting the molecular impact of genetic variation, yet have been criticized for poor performance in this task. Here we show that training models on functional genomic data with matched personal genomes improves their performance at variant effect prediction. Variant effect representations are retained even when fine tuning models to unseen cellular contexts and experimental readouts. Our results have implications for interpreting trait-associated genetic variation.
序列到功能模型在解释遗传变异的分子影响方面有广泛应用，但因在这项任务中表现不佳而受到批评。在这里，我们表明，在具有匹配个人基因组的功能基因组数据上训练模型，可以提高它们在变异效应预测方面的性能。即使在将模型微调至未见过的细胞背景和实验读数时，变异效应表征也能得以保留。我们的结果对解释与性状相关的遗传变异具有启示意义。
