Dataset Information

A rapid and reference-free imputation method for low-cost genotyping platforms.

ABSTRACT: Most current genotype imputation methods are reference-based, which posed several challenges to users, such as high computational costs and reference panel inaccessibility. Thus, deep learning models are expected to create reference-free imputation methods performing with higher accuracy and shortening the running time. We proposed a imputation method using recurrent neural networks integrating with an additional discriminator network, namely GRUD. This method was applied to datasets from genotyping chips and Low-Pass Whole Genome Sequencing (LP-WGS) with the reference panels from The 1000 Genomes Project (1KGP) phase 3, the dataset of 4810 Singaporeans (SG10K), and The 1000 Vietnamese Genome Project (VN1K). Our model performed more accurately than other existing methods on multiple datasets, especially with common variants with large minor allele frequency, and shrank running time and memory usage. In summary, these results indicated that GRUD can be implemented in genomic analyses to improve the accuracy and running-time of genotype imputation.

SUBMITTER: Chi Duong V

PROVIDER: S-EPMC10754833 | biostudies-literature | 2023 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A rapid and reference-free imputation method for low-cost genotyping platforms.

Chi Duong Vinh V Minh Vu Giang G Khac Nguyen Thien T Tran The Nguyen Hung H Luong Pham Thang T S Vo Nam N Hong Hoang Tham T

Scientific reports 20231227 1

Most current genotype imputation methods are reference-based, which posed several challenges to users, such as high computational costs and reference panel inaccessibility. Thus, deep learning models are expected to create reference-free imputation methods performing with higher accuracy and shortening the running time. We proposed a imputation method using recurrent neural networks integrating with an additional discriminator network, namely GRUD. This method was applied to datasets from genoty ...[more]

PMID: 38155188

Similar Datasets

Project description:Background: Genetic tests for primary immunodeficiency disorders (PIDs) are expensive, time-consuming, and not easily accessible in developing countries. Therefore, we studied the feasibility of a customized single nucleotide variant (SNV) microarray that we developed to detect disease-causing variants and copy number variation (CNV) in patients with PIDs for only 40 Euros. Methods: Probes were custom-designed to genotype 9,415 variants of 277 PID-related genes, and were added to the genome-wide Illumina Global Screening Array (GSA). Data analysis of GSA was performed using Illumina GenomeStudio 2.0, Biodiscovery Nexus 10.0, and R-3.4.4 software. Validation of genotype calling was performed by comparing the GSA with whole-genome sequencing (WGS) data of 56 non-PID controls. DNA samples of 95 clinically diagnosed PID patients, of which 60 patients (63%) had a genetically established diagnosis (by Next-Generation Sequencing (NGS) PID panels or Sanger sequencing), were analyzed to test the performance of the GSA. The additional SNVs detected by GSA were validated by Sanger sequencing. Results: Genotype calling of the customized array had an accuracy rate of 99.7%. The sensitivity for detecting rare PID variants was high (87%). The single sample replication in two runs was high (94.9%). The customized GSA was able to generate a genetic diagnosis in 37 out of 95 patients (39%). These 37 patients included 29 patients in whom the genetic variants were confirmed by conventional methods (26 patients by SNV and 3 by CNV analysis), while in 8 patients a new genetic diagnosis was established (6 patients by SNV and 2 patients suspected for leukemia by CNV analysis). Twenty-eight patients could not be detected due to the limited coverage of the custom probes. However, the diagnostic yield can potentially be increased when newly updated variants are added. Conclusion: Our robust customized GSA seems to be a promising first-line rapid screening tool for PIDs at an affordable price, which opens opportunities for low-cost genetic testing in developing countries. The technique is scalable, allows numerous new genetic variants to be added, and offers the potential for genetic testing not only in PIDs, but also in many other genetic diseases.

Dataset Information

A rapid and reference-free imputation method for low-cost genotyping platforms.

Publications

A rapid and reference-free imputation method for low-cost genotyping platforms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets