Transforming dbGaP genetic and genomic data to FAIR-ready by artificial intelligence and machine learning algorithms

通过人工智能和机器学习算法将 dbGaP 遗传和基因组数据转变为 FAIR-ready

基本信息

批准号：
10842954
负责人：
Zhongming Zhao
金额：
$ 30.61万
依托单位：
UNIVERSITY OF TEXAS HLTH SCI CTR HOUSTON
依托单位国家：
美国
项目类别：
财政年份：
2017
资助国家：
美国
起止时间：
2017-09-14 至 2025-05-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10842954
关键词：
ATAC-seq Acceleration Address Administrative Supplement Algorithmic Analysis Algorithms Alleles Alzheimer&apos s Disease Artificial Intelligence Attention Brain Brain Diseases Case Study Cells Chromosomes Code Cognitive Collection Complex Custom DNA sequencing DNase I hypersensitive sites sequencing Data Data Discovery Data Set Disease Funding Gene Expression Genes Genetic Genetic Markers Genetic Transformation Genetic study Genomics Genotype Glioblastoma Goals Grant Haplotypes Histones Human Image Image Analysis Impaired cognition Individual Investigation Language Logistic Regressions Machine Learning Methods Modeling Molecular Multiomic Data Multiple Sclerosis Mus Neural Network Simulation Pathway interactions Phenotype Principal Investigator Procedures Process Research Personnel Resolution Role Schizophrenia Screening procedure Standardization Testing Tissues Training United States National Institutes of Health United States National Library of Medicine Untranslated RNA Variant Vocabulary Weight Work artificial intelligence algorithm autoencoder cell type cognitive testing convolutional neural network database of Genotypes and Phenotypes deep learning deep learning algorithm deep learning model digital imaging dosage epigenomics gene discovery genetic resource genetic variant genome sequencing genome wide association study genomic data improved learning strategy machine learning algorithm malignant breast neoplasm multiple omics neural network algorithm phenotypic data programs prototype repository risk variant single-cell RNA sequencing spatiotemporal trait two-dimensional

项目摘要

dbGaP is a repository for NIH funded projects and it contains many genetic and genomic data. However, data there are not ready for AI and machine learning applications. This application proposes methods to address this issue. We have two aims: 1). Develop and standardize procedures to transfer genetic and genomic data into image like objects and tokenized custom vocabulary so that the data can be utilized by advanced AI algorithms such CNN, autoencoder and transformer. To transform genetic data into image, we recode allele dosage value as pixel intensity and arrange a collection of genetic markers such as SNPs and CNVs into an artificial image object so that it can be analyzed by CNN algorithms. Genetic markers can also be used to define haplotypes, which can be tokenized into custom vocabularies for use in NLP models. 2). Use Alzheimer's disease and schizophrenia as case studies to demonstrate the utilities of transformed data for the discovery and identification of risk variants/genes for both conditions. We plan to impute genetically controlled gene expression using brain specific eQTLs and individual genotypes for an AD dataset, and transform the expression data into image objects for analyses by CNN model with self attention mechanism. For schizophrenia, we plan to use k- mer tokenizer to break haplotypes into a collection of small haplotype blocks and treat them as tokens for analyses by NLP models. We use both CNN and NLP models as screen tools to select promising candidates using the attention weights, and then directly test these candidates for their association with AD/schizophrenia using logistic regression. Due to the selection effect, we can dramatically reduce the number of testing, significantly increase our statistical power to detect risk variants/genes to AD/schizophrenia.

dbGaP 是 NIH 资助项目的存储库，包含许多遗传和基因组数据。然而，那里的数据还没有准备好用于人工智能和机器学习应用。这个应用程序提出解决这个问题的方法。我们有两个目标：1）。开发和标准化将遗传和基因组数据传输到图像类对象和标记化定制的程序词汇表，以便数据可以被 CNN、自动编码器等高级 AI 算法利用和变压器。为了将遗传数据转换为图像，我们将等位基因剂量值重新编码为像素强度并将一系列遗传标记（例如 SNP 和 CNV）排列成人工图像对象，以便可以通过 CNN 算法对其进行分析。也可以使用遗传标记定义单倍型，可以将其标记为自定义词汇表以在 NLP 模型中使用。 2）。使用阿尔茨海默病和精神分裂症作为案例研究来证明转换数据以发现和识别这两种情况的风险变异/基因。我们计划使用大脑特定的 eQTL 来估算遗传控制的基因表达， AD 数据集的各个基因型，并将表达数据转换为图像对象通过具有自注意力机制的 CNN 模型进行分析。对于精神分裂症，我们计划使用 k- mer tokenizer 将单倍型分解为小单倍型块的集合并将它们视为用于 NLP 模型分析的标记。我们使用 CNN 和 NLP 模型作为筛选工具使用注意力权重选择有前途的候选者，然后直接测试这些候选者使用逻辑回归分析其与 AD/精神分裂症的关联。由于选择效应，我们可以大大减少测试数量，显着提高我们的统计能力检测 AD/精神分裂症的风险变异/基因。

项目成果

期刊论文数量（42）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Identification of gene signatures from RNA-seq data using Pareto-optimal cluster algorithm.

使用 Pareto 最优聚类算法从 RNA-seq 数据中识别基因特征。

DOI：
发表时间：
2018
期刊：
影响因子：
0
作者：
Mallik, Saurav;Zhao, Zhongming
通讯作者：
Zhao, Zhongming

Differential Expression of Viral Transcripts From Single-Cell RNA Sequencing of Moderate and Severe COVID-19 Patients and Its Implications for Case Severity.

中度和重症 COVID-19 患者单细胞 RNA 测序中病毒转录物的差异表达及其对病例严重程度的影响。

DOI：
发表时间：
2020
期刊：
影响因子：
0
作者：
Liu, Teng;Jia, Peilin;Fang, Bingliang;Zhao, Zhongming
通讯作者：
Zhao, Zhongming

Age-associated telomere attrition in adipocyte progenitors predisposes to metabolic disease.

脂肪细胞祖细胞中与年龄相关的端粒磨损易患代谢疾病。

DOI：
发表时间：
2020-12
期刊：
影响因子：
20.8
作者：
Gao, Zhanguo;Daquinag, Alexes C;Fussell, Cale;Zhao, Zhongming;Dai, Yulin;Rivera, Angielyn;Snyder, Brad E;Eckel;Kolonin, Mikhail G
通讯作者：
Kolonin, Mikhail G

Insulin resistance in depression: A large meta-analysis of metabolic parameters and variation.

抑郁症中的胰岛素抵抗：代谢参数和变化的大型荟萃分析。