Yield prediction through integration of genetic, environment, and management data through deep learning

Accurate prediction of the phenotypic outcomes produced by different combinations of genotypes, environments, and management interventions remains a key goal in biology with direct applications to agriculture, research, and conservation. The past decades have seen an expansion of new methods applied toward this goal. Here we predict maize yield using deep neural networks, compare the efficacy of 2 model development methods, and contextualize model performance using conventional linear and machine learning models. We examine the usefulness of incorporating interactions between disparate data types. We find deep learning and best linear unbiased predictor (BLUP) models with interactions had the best overall performance. BLUP models achieved the lowest average error, but deep learning models performed more consistently with similar average error. Optimizing deep neural network submodules for each data type improved model performance relative to optimizing the whole model for all data types at once. Examining the effect of interactions in the best-performing model revealed that including interactions altered the model's sensitivity to weather and management features, including a reduction of the importance scores for timepoints expected to have a limited physiological basis for influencing yield-those at the extreme end of the season, nearly 200 days post planting. Based on these results, deep learning provides a promising avenue for the phenotypic prediction of complex traits in complex environments and a potential mechanism to better understand the influence of environmental and genetic factors.

Maize Genomes to Fields: 2014 and 2015 field season genotype, phenotype, environment, and inbred ear image datasets

OBJECTIVES

Crop improvement relies on analysis of phenotypic, genotypic, and environmental data. Given large, well-integrated, multi-year datasets, diverse queries can be made: Which lines perform best in hot, dry environments? Which alleles of specific genes are required for optimal performance in each environment? Such datasets also can be leveraged to predict cultivar performance, even in uncharacterized environments. The maize Genomes to Fields (G2F) Initiative is a multi-institutional organization of scientists working to generate and analyze such datasets from existing, publicly available inbred lines and hybrids. G2F's genotype by environment project has released 2014 and 2015 datasets to the public, with 2016 and 2017 collected and soon to be made available.

DATA DESCRIPTION

Datasets include DNA sequences; traditional phenotype descriptions, as well as detailed ear, cob, and kernel phenotypes quantified by image analysis; weather station measurements; and soil characterizations by site. Data are released as comma separated value spreadsheets accompanied by extensive README text descriptions. For genotypic and phenotypic data, both raw data and a version with outliers removed are reported. For weather data, two versions are reported: a full dataset calibrated against nearby National Weather Service sites and a second calibrated set with outliers and apparent artifacts removed.

Influenza D virus M2 protein exhibits ion channel activity in Xenopus laevis oocytes

BACKGROUND

A new type of influenza virus, known as type D, has recently been identified in cattle and pigs. Influenza D virus infection in cattle is typically asymptomatic; however, its infection in swine can result in clinical disease. Swine can also be infected with all other types of influenza viruses, namely A, B, and C. Consequently, swine can serve as a "mixing vessel" for highly pathogenic influenza viruses, including those with zoonotic potential. Currently, the only antiviral drug available targets influenza M2 protein ion channel is not completely effective. Thus, it is necessary to develop an M2 ion channel blocker capable of suppressing the induction of resistance to the genetic shift. To provide a basis for developing novel ion channel-blocking compounds, we investigated the properties of influenza D virus M2 protein (DM2) as a drug target.

RESULTS

To test the ion channel activity of DM2, the DNA corresponding to DM2 with cMyc-tag conjugated to its carboxyl end was cloned into the shuttle vector pNCB1. The mRNA of the DM2-cMyc gene was synthesized and injected into Xenopus oocytes. The translation products of DM2-cMyc mRNA were confirmed by immunofluorescence and mass spectrometry analyses. The DM2-cMyc mRNA-injected oocytes were subjected to the two-electrode voltage-clamp (TEVC) method, and the induced inward current was observed. The midpoint (Vmid) values in Boltzmann modeling for oocytes injected with DM2-cMyc RNA or a buffer were -152 and -200 mV, respectively. Assuming the same expression level in the Xenopus oocytes, DM2 without tag and influenza C virus M2 protein (CM2) were subjected to the TEVC method. DM2 exhibited ion channel activity under the condition that CM2 ion channel activity was reproduced. The gating voltages represented by Vmid for CM2 and DM2 were -141 and -146 mV, respectively. The reversal potentials observed in ND96 for CM2 and DM2 were -21 and -22 mV, respectively. Compared with intact DM2, DM2 variants with mutation in the YxxxK motif, namely Y72A and K76A DM2, showed lower Vmid values while showing no change in reversal potential.

CONCLUSION

The M2 protein from newly isolated influenza D virus showed ion channel activity similar to that of CM2. The gating voltage was shown to be affected by the YxxxK motif and by the hydrophobicity and bulkiness of the carboxyl end of the molecule.

Resistant starch can improve insulin sensitivity independently of the gut microbiota

BACKGROUND

Obesity-related diseases, including type 2 diabetes and cardiovascular disease, have reached epidemic proportions in industrialized nations, and dietary interventions for their prevention are therefore important. Resistant starches (RS) improve insulin sensitivity in clinical trials, but the mechanisms underlying this health benefit remain poorly understood. Because RS fermentation by the gut microbiota results in the formation of physiologically active metabolites, we chose to specifically determine the role of the gut microbiota in mediating the metabolic benefits of RS. To achieve this goal, we determined the effects of RS when added to a Western diet on host metabolism in mice with and without a microbiota.

RESULTS

RS feeding of conventionalized mice improved insulin sensitivity and redressed some of the Western diet-induced changes in microbiome composition. However, parallel experiments in germ-free littermates revealed that RS-mediated improvements in insulin levels also occurred in the absence of a microbiota. RS reduced gene expression of adipose tissue macrophage markers and altered cecal concentrations of several bile acids in both germ-free and conventionalized mice; these effects were strongly correlated with the metabolic benefits, providing a potential microbiota-independent mechanism to explain the physiological effects of RS.

CONCLUSIONS

This study demonstrated that some metabolic benefits exerted by dietary RS, especially improvements in insulin levels, occur independently of the microbiota and could involve alterations in the bile acid cycle and adipose tissue immune modulation. This work also sets a precedent for future mechanistic studies aimed at establishing the causative role of the gut microbiota in mediating the benefits of bioactive compounds and functional foods.