Predicting and elucidating the etiology of fatty liver disease using a machine learning-based approach: an IMI DIRECT study-猫眼课题宝

DOI：

10.1101/2020.02.10.20021147

发表时间：

2020-02

期刊：

medRxiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

N. Atabaki-Pasdar;M. Ohlsson;A. Viñuela;F. Frau;H. Pomares-Millan;Mark Haid;A. Jones;Elizabeth Louise Thomas;R. Koivula;A. Kurbasic;Pascal M. Mutie;H. Fitipaldi;Juan Fernandez;A. Dawed;G. Giordano;I. Forgie;T. Mcdonald;F. Rutters;H. Cederberg;E. Chabanova;M. Dale;Federico De Masi;C. Thomas;K. Allin;T. Hansen;Alison J. Heggie;Mun-Gwan Hong;P. Elders;G. Kennedy;T. Kokkola;H. Pedersen;A. Mahajan;D. McEvoy;F. Pattou;V. Raverdy;Ragna S. Häussler;Sapna Sharma;H. Thomsen;J. Vangipurapu;H. Vestergaard;L. Hart;J. Adamski;P. Musholt;S. Brage;S. Brunak;E. Dermitzakis;G. Frost;T. Hansen;M. Laakso;O. Pedersen;M. Ridderstråle;H. Ruetten;A. Hattersley;M. Walker;J. Beulens;A. Mari;J. Schwenk;Ramneek Gupta;M. McCarthy;E. Pearson;Jimmy D Bell;I. Pávó;P. Franks

非酒精性脂肪性肝病（NAFLD）非常普遍，并导致2型糖尿病（T2D）及以上的严重健康并发症。NAFLD的早期诊断很重要，因为这可以帮助预防对肝脏的不可逆损伤和最终的肝细胞癌。利用IMI DIRECT参与者（n=1514）的基线数据，我们试图扩大病因学的理解，并使用机器学习开发NAFLD的诊断工具。多组学（遗传学、转录组学、蛋白质组学和代谢组学）和临床（肝酶和其他血清学生物标志物、人体测量学以及β细胞功能、胰岛素敏感性和生活方式的测量）数据构成关键输入变量。模型在MRI图像来源的肝脏脂肪含量（<5%或≥5%）上进行训练。我们应用LASSO（最小绝对收缩和选择算子）从组学数据的不同层中选择特征，并使用随机森林分析来开发模型。预测模型包括单独或组合的临床和组学变量。包括所有组学和临床变量的模型产生了0.84（95%置信区间（CI）=0.82，0.86）的交叉验证的受试者操作者特征曲线下面积（ROCAUC），而包括9个临床可访问变量的模型的ROCAUC为0.82（95% CI=0.81，0.83）。IMI DIRECT预测模型优于现有的非侵入性NAFLD预测工具。我们已经开发出临床上有用的肝脏脂肪预测模型（参见：www.predictliverfat.org），并确定了似乎影响肝脏脂肪积累的生物学特征。

Non-alcoholic fatty liver disease (NAFLD) is highly prevalent and causes serious health complications in type 2 diabetes (T2D) and beyond. Early diagnosis of NAFLD is important, as this can help prevent irreversible damage to the liver and ultimately hepatocellular carcinomas.Utilizing the baseline data from the IMI DIRECT participants (n=1514) we sought to expand etiological understanding and develop a diagnostic tool for NAFLD using machine learning. Multi-omic (genetic, transcriptomic, proteomic, and metabolomic) and clinical (liver enzymes and other serological biomarkers, anthropometry, and measures of beta-cell function, insulin sensitivity, and lifestyle) data comprised the key input variables. The models were trained on MRI image-derived liver fat content (<5% or ≥5%). We applied LASSO (least absolute shrinkage and selection operator) to select features from the different layers of omics data and Random Forest analysis to develop the models. The prediction models included clinical and omics variables separately or in combination. A model including all omics and clinical variables yielded a cross-validated receiver operator characteristic area under the curve (ROCAUC) of 0.84 (95% confidence interval (CI)=0.82, 0.86), which compared with a ROCAUC of 0.82 (95% CI=0.81, 0.83) for a model including nine clinically-accessible variables. The IMI DIRECT prediction models out-performed existing non-invasive NAFLD prediction tools.We have developed clinically useful liver fat prediction models (see:www.predictliverfat.org) and identified biological features that appear to affect liver fat accumulation.

Predicting and elucidating the etiology of fatty liver disease using a machine learning-based approach: an IMI DIRECT study

文献摘要

相似文献