A genotype imputation method for de-identified haplotype reference information by using recurrent neural network

STICI: Split-Transformer with integrated convolutions for genotype imputation

Despite advances in sequencing technologies, genome-scale datasets often contain missing bases and genomic segments, hindering downstream analyses. Genotype imputation addresses this issue and has been a cornerstone pre-processing step in genetic and genomic studies. Although various methods have been widely adopted for genotype imputation, it remains challenging to impute certain genomic regions and large structural variants. Here, we present a transformer-based framework, named STICI, for accurate genotype imputation. STICI models automatically learn genome-wide patterns of linkage disequilibrium, evidenced by much higher imputation accuracy in regions with highly linked variants. Our imputation results on the human 1000 Genomes Project and non-human genomes show that STICI can achieve high imputation accuracy comparable to the state-of-the-art genotype imputation methods, with the additional capability to impute multi-allelic variants and various types of genetic variants. STICI can be trained for any collection of genomes automatically using self-supervision. Moreover, STICI shows excellent performance without needing any special presuppositions about the underlying patterns in collections of non-human genomes, pointing to adaptability and applications of STICI to impute missing genotypes in any species.

Genotype imputation methods for whole and complex genomic regions utilizing deep learning technology

The imputation of unmeasured genotypes is essential in human genetic research, particularly in enhancing the power of genome-wide association studies and conducting subsequent fine-mapping. Recently, several deep learning-based genotype imputation methods for genome-wide variants with the capability of learning complex linkage disequilibrium patterns have been developed. Additionally, deep learning-based imputation has been applied to a distinct genomic region known as the major histocompatibility complex, referred to as HLA imputation. Despite their various advantages, the current deep learning-based genotype imputation methods do have certain limitations and have not yet become standard. These limitations include the modest accuracy improvement over statistical and conventional machine learning-based methods. However, their benefits include other aspects, such as their "reference-free" nature, which ensures complete privacy protection, and their higher computational efficiency. Furthermore, the continuing evolution of deep learning technologies is expected to contribute to further improvements in prediction accuracy and usability in the future.

Two-stage strategy using denoising autoencoders for robust reference-free genotype imputation with missing input genotypes

Widely used genotype imputation methods are based on the Li and Stephens model, which assumes that new haplotypes can be represented by modifying existing haplotypes in a reference panel through mutations and recombinations. These methods use genotypes from SNP arrays as inputs to estimate haplotypes that align with the input genotypes by analyzing recombination patterns within a reference panel, and then infer unobserved variants. While these methods require reference panels in an identifiable form, their public use is limited due to privacy and consent concerns. One strategy to overcome these limitations is to use de-identified haplotype information, such as summary statistics or model parameters. Advances in deep learning (DL) offer the potential to develop imputation methods that use haplotype information in a reference-free manner by handling it as model parameters, while maintaining comparable imputation accuracy to methods based on the Li and Stephens model. Here, we provide a brief introduction to DL-based reference-free genotype imputation methods, including RNN-IMP, developed by our research group. We then evaluate the performance of RNN-IMP against widely-used Li and Stephens model-based imputation methods in terms of accuracy (R2), using the 1000 Genomes Project Phase 3 dataset and corresponding simulated Omni2.5 SNP genotype data. Although RNN-IMP is sensitive to missing values in input genotypes, we propose a two-stage imputation strategy: missing genotypes are first imputed using denoising autoencoders; RNN-IMP then processes these imputed genotypes. This approach restores the imputation accuracy that is degraded by missing values, enhancing the practical use of RNN-IMP.

Efficient HLA imputation from sequential SNPs data by transformer

Human leukocyte antigen (HLA) genes are associated with a variety of diseases, yet the direct typing of HLA alleles is both time-consuming and costly. Consequently, various imputation methods leveraging sequential single nucleotide polymorphisms (SNPs) data have been proposed, employing either statistical or deep learning models, such as the convolutional neural network (CNN)-based model, DEEP*HLA. However, these methods exhibit limited imputation efficiency for infrequent alleles and necessitate a large size of reference dataset. In this context, we have developed a Transformer-based model to HLA allele imputation, named "HLA Reliable IMpuatioN by Transformer (HLARIMNT)" designed to exploit the sequential nature of SNPs data. We evaluated HLARIMNT's performance using two distinct reference panels; Pan-Asian reference panel (n = 530) and Type 1 Diabetes genetics Consortium (T1DGC) reference panel (n = 5225), alongside a combined panel (n = 1060). HLARIMNT demonstrated superior accuracy to DEEP*HLA across several indices, particularly for infrequent alleles. Furthermore, we explored the impact of varying training data sizes on imputation accuracy, finding that HLARIMNT consistently outperformed across all data size. These findings suggest that Transformer-based models can efficiently impute not only HLA types but potentially other gene types from sequential SNPs data.

Predictive ability of multi-population genomic prediction methods of phenotypes for reproduction traits in Chinese and Austrian pigs

BACKGROUND

Multi-population genomic prediction can rapidly expand the size of the reference population and improve genomic prediction ability. Machine learning (ML) algorithms have shown advantages in single-population genomic prediction of phenotypes. However, few studies have explored the effectiveness of ML methods for multi-population genomic prediction.

RESULTS

In this study, 3720 Yorkshire pigs from Austria and four breeding farms in China were used, and single-trait genomic best linear unbiased prediction (ST-GBLUP), multitrait GBLUP (MT-GBLUP), Bayesian Horseshoe (BayesHE), and three ML methods (support vector regression (SVR), kernel ridge regression (KRR) and AdaBoost.R2) were compared to explore the optimal method for joint genomic prediction of phenotypes of Chinese and Austrian pigs through 10 replicates of fivefold cross-validation. In this study, we tested the performance of different methods in two scenarios: (i) including only one Austrian population and one Chinese pig population that were genetically linked based on principal component analysis (PCA) (designated as the "two-population scenario") and (ii) adding reference populations that are unrelated based on PCA to the above two populations (designated as the "multi-population scenario"). Our results show that, the use of MT-GBLUP in the two-population scenario resulted in an improvement of 7.1% in predictive ability compared to ST-GBLUP, while the use of SVR and KKR yielded improvements in predictive ability of 4.5 and 5.3%, respectively, compared to MT-GBLUP. SVR and KRR also yielded lower mean square errors (MSE) in most population and trait combinations. In the multi-population scenario, improvements in predictive ability of 29.7, 24.4 and 11.1% were obtained compared to ST-GBLUP when using, respectively, SVR, KRR, and AdaBoost.R2. However, compared to MT-GBLUP, the potential of ML methods to improve predictive ability was not demonstrated.

CONCLUSIONS

Our study demonstrates that ML algorithms can achieve better prediction performance than multitrait GBLUP models in multi-population genomic prediction of phenotypes when the populations have similar genetic backgrounds; however, when reference populations that are unrelated based on PCA are added, the ML methods did not show a benefit. When the number of populations increased, only MT-GBLUP improved predictive ability in both validation populations, while the other methods showed improvement in only one population.

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

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.

Rapid, Reference-Free human genotype imputation with denoising autoencoders

Genotype imputation is a foundational tool for population genetics. Standard statistical imputation approaches rely on the co-location of large whole-genome sequencing-based reference panels, powerful computing environments, and potentially sensitive genetic study data. This results in computational resource and privacy-risk barriers to access to cutting-edge imputation techniques. Moreover, the accuracy of current statistical approaches is known to degrade in regions of low and complex linkage disequilibrium. Artificial neural network-based imputation approaches may overcome these limitations by encoding complex genotype relationships in easily portable inference models. Here, we demonstrate an autoencoder-based approach for genotype imputation, using a large, commonly used reference panel, and spanning the entirety of human chromosome 22. Our autoencoder-based genotype imputation strategy achieved superior imputation accuracy across the allele-frequency spectrum and across genomes of diverse ancestry, while delivering at least fourfold faster inference run time relative to standard imputation tools.

Construction of a trio-based structural variation panel utilizing activated T lymphocytes and long-read sequencing technology

Long-read sequencing technology enable better characterization of structural variants (SVs). To adapt the technology to population-scale analyses, one critical issue is to obtain sufficient amount of high-molecular-weight genomic DNA. Here, we propose utilizing activated T lymphocytes, which can be established efficiently in a biobank to stably supply high-grade genomic DNA sufficiently. We conducted nanopore sequencing of 333 individuals constituting 111 trios with high-coverage long-read sequencing data (depth 22.2x, N50 of 25.8 kb) and identified 74,201 SVs. Our trio-based analysis revealed that more than 95% of the SVs were concordant with Mendelian inheritance. We also identified SVs associated with clinical phenotypes, all of which appear to be stably transmitted from parents to offspring. Our data provide a catalog of SVs in the general Japanese population, and the applied approach using the activated T-lymphocyte resource will contribute to biobank-based human genetic studies focusing on SVs at the population scale.

Long-read sequencing on activated T-cells from a sample of 333 Japanese individuals (representing 111 parent-offspring trios) provides a useful reference dataset of structural variation in the Japanese population.

Evaluation of vicinity-based hidden Markov models for genotype imputation

BACKGROUND

The decreasing cost of DNA sequencing has led to a great increase in our knowledge about genetic variation. While population-scale projects bring important insight into genotype-phenotype relationships, the cost of performing whole-genome sequencing on large samples is still prohibitive. In-silico genotype imputation coupled with genotyping-by-arrays is a cost-effective and accurate alternative for genotyping of common and uncommon variants. Imputation methods compare the genotypes of the typed variants with the large population-specific reference panels and estimate the genotypes of untyped variants by making use of the linkage disequilibrium patterns. Most accurate imputation methods are based on the Li-Stephens hidden Markov model, HMM, that treats the sequence of each chromosome as a mosaic of the haplotypes from the reference panel.

RESULTS

Here we assess the accuracy of vicinity-based HMMs, where each untyped variant is imputed using the typed variants in a small window around itself (as small as 1 centimorgan). Locality-based imputation is used recently by machine learning-based genotype imputation approaches. We assess how the parameters of the vicinity-based HMMs impact the imputation accuracy in a comprehensive set of benchmarks and show that vicinity-based HMMs can accurately impute common and uncommon variants.

CONCLUSIONS

Our results indicate that locality-based imputation models can be effectively used for genotype imputation. The parameter settings that we identified can be used in future methods and vicinity-based HMMs can be used for re-structuring and parallelizing new imputation methods. The source code for the vicinity-based HMM implementations is publicly available at https://github.com/harmancilab/LoHaMMer .

Using machine learning to improve the accuracy of genomic prediction of reproduction traits in pigs

Background

Recently, machine learning (ML) has become attractive in genomic prediction, but its superiority in genomic prediction over conventional (ss) GBLUP methods and the choice of optimal ML methods need to be investigated.

Results

In this study, 2566 Chinese Yorkshire pigs with reproduction trait records were genotyped with the GenoBaits Porcine SNP 50 K and PorcineSNP50 panels. Four ML methods, including support vector regression (SVR), kernel ridge regression (KRR), random forest (RF) and Adaboost.R2 were implemented. Through 20 replicates of fivefold cross-validation (CV) and one prediction for younger individuals, the utility of ML methods in genomic prediction was explored. In CV, compared with genomic BLUP (GBLUP), single-step GBLUP (ssGBLUP) and the Bayesian method BayesHE, ML methods significantly outperformed these conventional methods. ML methods improved the genomic prediction accuracy of GBLUP, ssGBLUP, and BayesHE by 19.3%, 15.0% and 20.8%, respectively. In addition, ML methods yielded smaller mean squared error (MSE) and mean absolute error (MAE) in all scenarios. ssGBLUP yielded an improvement of 3.8% on average in accuracy compared to that of GBLUP, and the accuracy of BayesHE was close to that of GBLUP. In genomic prediction of younger individuals, RF and Adaboost.R2_KRR performed better than GBLUP and BayesHE, while ssGBLUP performed comparably with RF, and ssGBLUP yielded slightly higher accuracy and lower MSE than Adaboost.R2_KRR in the prediction of total number of piglets born, while for number of piglets born alive, Adaboost.R2_KRR performed significantly better than ssGBLUP. Among ML methods, Adaboost.R2_KRR consistently performed well in our study. Our findings also demonstrated that optimal hyperparameters are useful for ML methods. After tuning hyperparameters in CV and in predicting genomic outcomes of younger individuals, the average improvement was 14.3% and 21.8% over those using default hyperparameters, respectively.

Conclusion

Our findings demonstrated that ML methods had better overall prediction performance than conventional genomic selection methods, and could be new options for genomic prediction. Among ML methods, Adaboost.R2_KRR consistently performed well in our study, and tuning hyperparameters is necessary for ML methods. The optimal hyperparameters depend on the character of traits, datasets etc.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40104-022-00708-0.

Application of Genomic Big Data in Plant Breeding:Past, Present, and Future

Plant breeding has a long history of developing new varieties that have ensured the food security of the human population. During this long journey together with humanity, plant breeders have successfully integrated the latest innovations in science and technologies to accelerate the increase in crop production and quality. For the past two decades, since the completion of human genome sequencing, genomic tools and sequencing technologies have advanced remarkably, and adopting these innovations has enabled us to cost down and/or speed up the plant breeding process. Currently, with the growing mass of genomic data and digitalized biological data, interdisciplinary approaches using new technologies could lead to a new paradigm of plant breeding. In this review, we summarize the overall history and advances of plant breeding, which have been aided by plant genomic research. We highlight the key advances in the field of plant genomics that have impacted plant breeding over the past decades and introduce the current status of innovative approaches such as genomic selection, which could overcome limitations of conventional breeding and enhance the rate of genetic gain.