Personal transcriptome variation is poorly explained by current genomic deep learning models.
Personal transcriptome variation is poorly explained by current genomic deep learning models.
复制标题
DOI:
10.1038/s41588-023-01574-w
复制
发表时间:
2023-12
期刊:
影响因子:
30.8
通讯作者:
Ioannidis, Nilah M.
中科院分区:
文献类型:
--
作者:
Huang, Connie;Shuai, Richard W.;Baokar, Parth;Chung, Ryan;Rastogi, Ruchir;Kathail, Pooja;Ioannidis, Nilah M.
Genomic deep learning models can predict genome-wide epigenetic features and gene expression levels directly from DNA sequence. While current models perform well at predicting gene expression levels across genes in different cell types from the reference genome, their ability to explain expression variation between individuals due to cis-regulatory genetic variants remains largely unexplored. Here, we evaluate four state-of-the-art models on paired personal genome and transcriptome data and find limited performance when explaining variation in expression across individuals. In addition, models often fail to predict the correct direction of effect of cis-regulatory genetic variation on expression. A test of four genomic sequence-to-expression deep learning models (Enformer, Basenji2, ExPecto, Xpresso) finds that they often fail to predict the correct direction of effect of cis-regulatory genetic variation on gene expression.
登录
查看更多内容
影响因子:
64.8
作者:
通讯作者:
--
影响因子:
8.8
作者:
Agarwal, Vikram;Shendure, Jay
通讯作者:
Shendure, Jay
影响因子:
30.8
作者:
Zhou, Jian
通讯作者:
Zhou, Jian
影响因子:
16.6
作者:
Wang QS;Kelley DR;Ulirsch J;Kanai M;Sadhuka S;Cui R;Albors C;Cheng N;Okada Y;Biobank Japan Project;Aguet F;Ardlie KG;MacArthur DG;Finucane HK
通讯作者:
Finucane HK
影响因子:
30.8
作者:
Chen, Kathleen M.;Wong, Aaron K.;Troyanskaya, Olga G.;Zhou, Jian
通讯作者:
Zhou, Jian