RESUMO
With the improvement of high-throughput technologies in recent years, large multi-dimensional plant omics data have been produced, and big-data-driven yield prediction research has received increasing attention. Machine learning offers promising computational and analytical solutions to interpret the biological meaning of large amounts of data in crops. In this study, we utilized multi-omics datasets from 156 maize recombinant inbred lines, containing 2496 single nucleotide polymorphisms (SNPs), 46 image traits (i-traits) from 16 developmental stages obtained through an automatic phenotyping platform, and 133 primary metabolites. Based on benchmark tests with different types of prediction models, some machine learning methods, such as Partial Least Squares (PLS), Random Forest (RF), and Gaussian process with Radial basis function kernel (GaussprRadial), achieved better prediction for maize yield, albeit slight difference for method preferences among i-traits, genomic, and metabolic data. We found that better yield prediction may be caused by various capabilities in ranking and filtering data features, which is found to be linked with biological meaning such as photosynthesis-related or kernel development-related regulations. Finally, by integrating multiple omics data with the RF machine learning approach, we can further improve the prediction accuracy of grain yield from 0.32 to 0.43. Our research provides new ideas for the application of plant omics data and artificial intelligence approaches to facilitate crop genetic improvements. Supplementary Information: The online version contains supplementary material available at 10.1007/s11032-024-01454-z.
RESUMO
BACKGROUND: Epidemiological and experimental evidences have implicated chronic inflammation in the association with allergic rhinitis (AR). However, it remains unclear whether specific circulating cytokines are the cause of AR or the consequence of bias. To examine whether genetic-predicted changes in circulating cytokine concentrations are related to the occurrence of AR, we conducted a two-sample Mendelian randomization (MR) analysis. METHODS: We investigated the causal effects of 26 circulating inflammatory cytokines on AR through MR analysis. The primary method employed in this study was the inverse variance-weighted (IVW) method. Sensitivity analyses were conducted using simple median, weighted median, penalized weighted median, and MR-Egger regression. RESULTS: Our study revealed suggestive evidence that higher levels of circulating IL-18 (OR per one standard deviation [SD] increase: 1.006; 95 % CI, 1.002 to 1.011; P = 0.006, PFDR = 0.067, random-effects IVW method) and Macrophage inflammatory protein-1α (MIP-1α) (OR per one SD increase: 1.015; 95 % CI, 1.004 to 1.026; P = 0.009, PFDR = 0.048, random-effects IVW method) were associated with an increased risk of AR. Conversely, higher levels of circulating TRAIL were associated with a decreased risk of AR (OR per one SD increase: 0.993; 95 % CI, 0.989 to 0.997; P = 4.58E-4, PFDR = 0.004, random-effects IVW method). Only the results of TRAIL exist after Bonferroni-correction (the p-value < 0.0019). Sensitivity analysis yielded directionally consistent results. No significant associations were observed between other circulating inflammatory cytokines and AR. CONCLUSION: Genetically predicted levels of IL-18, and MIP-1α are likely to associated with an increased risk of AR occurrence. Genetically predicted levels of TRAIL are statistically significant in reducing the risk of AR occurrence. However, the current research evidence does not support an impact of other inflammatory cytokines on the risk of AR. Future studies are needed to provide additional evidence to support the current conclusions.