Monitoring genetic gain is critical for evaluating breeding program performance. This study assessed genetic trends in the Zambia national maize breeding program using historical data (2001-2017) from 2225 hybrids tested across years and locations. Best linear unbiased estimates (BLUEs) were calculated, and genetic trends were determined by regressing entry means on first-year testing data. Mean heritability was moderate for grain yield, plant height, and ear height, and high for anthesis and silking dates, indicating strong reliability for flowering traits. Significant positive genetic gains were observed for most traits except days to silking. Grain yield (GY) increased at 0.021 t ha-1 per year (0.85% annually), reflecting progress but remaining below levels required to meet regional future production demands. Plant and ear height increased by more than 1.3 cm annually, suggesting directional selection for taller plant architecture. Grain texture declined by 1.28% per year, indicating a shift toward flint-type kernels. Anthesis date and ears per plant showed minimal genetic variation. Regression models explained more than 15% of the total variation in plant height, ear height, ear number, and grain texture, confirming consistent genetic progress. Although measurable gains were achieved, the study's baseline indicates that accelerating yield improvement will require rapid-cycle breeding, enhanced trait heritability, modern breeding tools, and a strategic reallocation of resources to sustain long-term impact.

Spatial variation is a major source of error in agricultural field experiments affecting genotype performance prediction. Implementing statistical models that account for spatial effects can improve the prediction of genotype performance. This study evaluated the impact of the P-spline spatial correction method on the estimation of genetic parameters and AIC values in two distinct crops, wheat and cassava, using four models: Block, Block + Spatial, Block + Marker, and Block + Marker + Spatial. Analyses were performed on data from 115 and 68 trials obtained from the T3/WheatCAP and Cassavabase databases, respectively. As assessed using Cullis heritability estimates and AIC values, the results demonstrated that correcting for spatial variation improved analyses of grain yield, test weight, plant height, powdery mildew, stripe rust, and bacterial streak disease in wheat. Similar improvements were observed in cassava for dry matter content, dry yield, and plant height. However, no improvement was observed for cassava mosaic disease or bacterial blight. These results were consistent whether or not marker effects were fitted in the models. This study demonstrates that incorporating spatial correction into statistical analyses substantially improves the precision of variety evaluation. By accounting for field heterogeneity, spatial modeling complements experimental design and enhances the accuracy of treatment comparisons. Therefore, integrating robust experimental designs with appropriate spatial analyses is essential for achieving optimal precision and reliability in field trial evaluations.

Increasing the rate of genetic gain in breeding programs is a critical component of crop genetic improvement strategies to increase yields in smallholder farmers' fields. While a growing array of technologies and tools are being deployed within breeding programs, optimizing resource allocation could provide a simple yet effective way to increase genetic gain, particularly within resource-constrained breeding programs. The objective of this study was to demonstrate that an easy-to-use deterministic model and a breeding costing tool could identify key modifications to improve the efficiency of breeding within the Zimbabwean national maize breeding program. The current program uses pedigree inbreeding, with a 4-1-1 tester scheme, and relatively low selection intensity. The method of inbreeding, test-crossing schemes, and selection intensity were modified within the current program budget. A combination of using doubled haploid lines, a 2-2-1 tester plan, and increased selection intensity improved gain per cycle by 42.8%, gain per year by 161.8%, gain per dollar by 43.1%, and decreased cost of one unit of genetic gain by 28.5% without a change in budget. Our results highlight how a simple deterministic model can identify steps to greatly improve breeding efficiency within resource-constrained breeding programs.

Assessing genetic gains from historical data provides insights to improve breeding programs. This study evaluated the Mozambique National Maize Program’s (MNMP’s) genetic gains using data from advanced germplasm trials conducted at 21 locations between 2014 and 2020. Genetic gains were calculated by regressing the genotypic best linear unbiased estimates of grain yield and complementary agronomic traits against the initial year of genotype evaluation (n = 592). The annual genetic gain was expressed as a percentage of the trait mean. While grain yield, the primary breeding focus, showed no significant improvement, significant gains were observed for the plant height (0.67%), ear height (1.74%), ears per plant (1.31%), ear position coefficient (1.22%), and husk cover (4.7%). Negative genetic gains were detected for the days to anthesis (−0.5%), the anthesis–silking interval or ASI (−9.31%), and stalk lodging (−5.01%). These results indicate that while MNMP did not achieve the desired positive genetic gain for grain yield, progress was made for traits related to plant resilience, particularly the ASI and stalk lodging. MNMP should seek to incorporate new breeding technologies and human resources to enhance genetic gains for grain yield and other key traits in the maize breeding program, while developing and deploying high-yielding, climate-resilient maize varieties to address emerging food security challenges in Mozambique.

This study explores the use of leaf-level visible-to-shortwave infrared (VSWIR) reflectance observations and partial least squares regression (PLSR) to predict foliar concentrations of macronutrients (nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur), micronutrients (boron, copper, iron, manganese, zinc, molybdenum, aluminum, and sodium), and moisture content in winter wheat. A total of 360 fresh wheat leaf samples were collected from a wheat breeding population over two growing seasons. These leaf samples were used to collect VSWIR reflectance observations across a spectral range spanning 350 to 2,500 nm. These samples were then processed for nutrient composition to allow for the examination of the ability of reflectance to accurately model diverse chemical components in wheat foliage. Models for each nutrient were developed using a rigorous cross-validation methodology in conjunction with three distinct component selection methods to explore the trade-offs between model complexity and performance in the final models. We examined absolute minimum predicted residual error sum of squares (PRESS), backward iteration over PRESS, and Van der Voet’s randomized t-test as component selection methods. In addition to contrasting component selection methods for each leaf trait, the importance of spectral regions through variable importance in projection scores was also examined. In general, the backward iteration method provided strong model performance while reducing model complexity relative to the other selection methods, yielding R2 [relative percent difference (RPD), root mean squared error (RMSE)] values in the validation dataset of 0.84 (2.45, 6.91), 0.75 (1.97, 18.67), 0.78 (2.13, 16.49), 0.66 (1.71, 17.13), 0.68 (1.75, 14.51), 0.66 (1.72, 12.29), and 0.84 (2.46, 2.20) for nitrogen, calcium, magnesium, sulfur, iron, zinc, and moisture content on a wet basis, respectively. These model results demonstrate that VSWIR reflectance in combination with modern statistical modeling techniques provides a powerful high throughput method for the quantification of a wide range of foliar nutrient contents in wheat crops. This work has the potential to advance rapid, precise, and nondestructive field assessments of nutrient contents and deficiencies for precision agricultural management and to advance breeding program assessments.

Monitoring genetic gains within breeding programs is a critical component for continuous improvement. While several national breeding programs in Africa have assessed genetic gain using era studies, this study is the first to use two decades of historical data to estimate genetic trends within a national breeding program. The objective of this study was to assess genetic trends within the final two stages of Zimbabwe’s Department of Research & Specialist Services maize breeding pipeline between 2002 and 2021. Data from 107 intermediate and 162 advanced variety trials, comprising of 716 and 398 entries, respectively, was analyzed. Trials were conducted under optimal, managed drought stress, low nitrogen stress, low pH, random stress, and disease pressure (maize streak virus (MSV), grey leaf spot (GLS), and turcicum leaf blight under artificial inoculation. There were positive and significant genetic gains for grain yield across management conditions (28–35 kg ha-1 yr-1), under high-yield potential environments (17–61 kg ha-1 yr-1), and under low-yield potential environments (0–16 kg ha-1 yr-1). No significant changes were observed in plant and ear height over the study period. Stalk and root lodging, as well as susceptibility to MSV and GLS, significantly decreased over the study period. New breeding technologies need to be incorporated into the program to further increase the rate of genetic gain in the maize breeding programs and to effectively meet future needs.

Optimization of a breeding program requires assessing and quantifying empirical genetic trends made through past efforts relative to the current breeding strategies, germplasm, technologies, and policy. To establish the genetic trends in the Kenyan Highland Maize Breeding Program (KHMP), a two-decade (1999-2020) historical dataset from the Preliminary Variety Trials (PVT) and Advanced Variety Trials (AVT) was analyzed. A mixed model analysis was used to compute the genetic gains for traits based on the best linear unbiased estimates in the PVT and AVT evaluation stages. A positive significant genetic gain estimate for grain yield of 88 kg ha-1 year-1 (1.94% year-1) and 26 kg ha-1 year-1 (0.42% year-1) was recorded for PVT and AVT, respectively. Root lodging, an important agronomic trait in the Kenya highlands, had a desired genetic gain of -2.65% year-1 for AVT. Results showed improvement in resistance to Turcicum Leaf Blight (TLB) with -1.19% and -0.27% year-1 for the PVT and AVT, respectively. Similarly, a significant genetic trend of -0.81% was noted for resistance to Gray Leaf Spot (GLS) in AVT. These findings highlight the good progress made by KHMP in developing adapted maize hybrids for Kenya's highland agroecology. Nevertheless, the study identified significant opportunities for the KHMP to make even greater genetic gains for key traits with introgression of favorable alleles for various traits, implementing a continuous improvement plan including marker-assisted forward breeding, sparse testing, and genomic selection, and doubled haploid technology for line development.

Estimating genetic gains is vital to optimize breeding programs for increased efficiency. Genetic gains should translate into productivity gains if returns to investments in breeding and impact are to be realized. The objective of this study was to estimate genetic gain for grain yield and key agronomic traits in pre-commercial and commercial maize varieties from public and private breeding programs tested in (i) national performance trials (NPT), (ii) era trial and, (iii) compare the trends with the national average. The study used (i) historical NPT data on 419 improved maize varieties evaluated in 23 trials at 6-8 locations each between 2008 and 2020, and (ii) data from an era trial of 54 maize hybrids released between 1999 and 2020. The NPT data was first analyzed using a mixed model and resulting estimate for each entry was regressed onto its first year of testing. Analysis was done over all entries, only entries from National Agricultural Research Organization (NARO), International Maize and Wheat Improvement Center (CIMMYT), or private seed companies. Estimated genetic gain was 2.25% or 81 kg ha-1 year-1 from the NPT analysis. A comparison of genetic trends by source indicated that CIMMYT entries had a gain of 1.98% year-1 or 106 kg ha-1 year-1. In contrast, NARO and private sector maize entries recorded genetic gains of 1.30% year-1 (59 kg ha-1 year-1) and 1.71% year-1 (79 kg ha-1 year-1), respectively. Varieties from NARO and private sector showed comparable mean yields of 4.56 t ha-1 and 4.62 t ha-1, respectively, while hybrids from CIMMYT had a mean of 5.37 t ha-1. Era analysis indicated significant genetic gain of 1.69% year-1 or 55 kg ha-1 year-1, while a significant national productivity gain of 1.48% year-1 (37 kg ha-1 year-1) was obtained. The study, thus, demonstrated the importance of public-private partnerships in development and delivery of new genetics to farmers in Uganda.

We documented changes in the wheat genome attributed to genomic selection including loss of diversity, and changes in population structure and linkage disequilibrium patterns. We conclude that training and prediction populations need to co-evolve instead of the use of a static training population. Genomic selection (GS) is widely used in plant breeding to shorten breeding cycles. Our objective was to assess the impact of rapid cycling GS on the wheat genome. We used 3927 markers to genotype a training population (YTP) and individuals from five cycles (YC1–YC5) of GS for grain yield. We assessed changes of allele frequency, genetic distance, population structure, and linkage disequilibrium (LD). We found 27.3

Fusarium head blight (FHB) is a devastating disease of wheat and barley. In the U.S.A., a significant long-term investment in breeding FHB-resistant cultivars began after the 1990s. However, to this date, no study has been performed to understand and monitor the rate of genetic progress in FHB resistance as a result of this investment. Using 20 years of data (1998 to 2018) from the Northern Uniform and Preliminarily Northern Uniform winter wheat scab nurseries that consisted of 1,068 genotypes originating from nine different institutions, we studied the genetic trends in FHB resistance within the northern soft red winter wheat growing region using mixed model analyses. For the FHB resistance traits incidence, severity, Fusarium-damaged kernels, and deoxynivalenol content, the rate of genetic gain in disease resistance was estimated to be 0.30 ± 0.1, 0.60 ± 0.09, and 0.37 ± 0.11 points per year, and 0.11 ± 0.05 parts per million per year, respectively. Among the five FHB-resistance quantitative trait loci assayed for test entries from 2012 to 2018, the frequencies of favorable alleles from Fhb 2DL Wuhan1 W14, Fhb Ernie 3Bc, and Fhb 5A Ning7840 were close to zero across the years. The frequency of the favorable at Fhb1 and Fhb 5A Ernie ranged from 0.08 to 0.33 and 0.06 to 0.20, respectively, across years, and there was no trend in changes in allele frequencies over years. Overall, this study showed that substantial genetic progress has been made toward improving resistance to FHB. It is apparent that today’s investment in public wheat breeding for FHB resistance is achieving results and will continue to play a vital role in reducing FHB levels in growers’ fields.

Being one of the most important staple dietary constituents globally, genetic enhancement of cultivated rice for yield, agronomically important traits is of substantial importance. Even though the climatic factors and crop management practices impact complex traits like yield immensely, the contribution of variation by underlying genetic factors surpasses them all. Previous studies have highlighted the importance of utilizing exotic germplasm, landraces in enhancing the diversity of gene pool, leading to better selections and thus superior cultivars. Thus, to fully exploit the potential of progenitor of Asian cultivated rice for productivity related traits, genome wide association study (GWAS) for seven agronomically important traits was conducted on a panel of 346 O. rufipogon accessions using a set of 15,083 high-quality single nucleotide polymorphic markers. The phenotypic data analysis indicated large continuous variation for all the traits under study, with a significant negative correlation observed between grain parameters and agronomic parameters like plant height, culm thickness. The presence of 74.28% admixtures in the panel as revealed by investigating population structure indicated the panel to be very poorly genetically differentiated, with rapid LD decay. The genome-wide association analyses revealed a total of 47 strong MTAs with 19 SNPs located in/close to previously reported QTL/genic regions providing a positive analytic proof for our studies. The allelic differences of significant MTAs were found to be statistically significant at 34 genomic regions. A total of 51 O. rufipogon accessions harboured combination of superior alleles and thus serve as potential candidates for accelerating rice breeding programs. The present study identified 27 novel SNPs to be significantly associated with different traits. Allelic differences between cultivated and wild rice at significant MTAs determined superior alleles to be absent at 12 positions implying substantial scope of improvement by their targeted introgression into cultivars. Introgression of novel significant genomic regions into breeder's pool would broaden the genetic base of cultivated rice, thus making the crop more resilient.

Additional file 2: Genotypic data of studied 346 accessions.

Striga hermonthica is a widespread, destructive parasitic plant that causes substantial yield loss to maize productivity in sub-Saharan Africa. Under severe Striga infestation, yield losses can range from 60 to 100% resulting in abandonment of farmers' lands. Diverse methods have been proposed for Striga management; however, host plant resistance is considered the most effective and affordable to small-scale famers. Thus, conducting a genome-wide association study to identify quantitative trait nucleotides controlling S. hermonthica resistance and mining of relevant candidate genes will expedite the improvement of Striga resistance breeding through marker-assisted breeding. For this study, 150 diverse maize inbred lines were evaluated under Striga infested and non-infested conditions for two years and genotyped using the genotyping-by-sequencing platform. Heritability estimates of Striga damage ratings, emerged Striga plants and grain yield, hereafter referred to as Striga resistance-related traits, were high under Striga infested condition. The mixed linear model (MLM) identified thirty SNPs associated with the three Striga resistance-related traits based on the multi-locus approaches (mrMLM, FASTmrMLM, FASTmrEMMA and pLARmEB). These SNPs explained up to 14% of the total phenotypic variation. Under non-infested condition, four SNPs were associated with grain yield, and these SNPs explained up to 17% of the total phenotypic variation. Gene annotation of significant SNPs identified candidate genes (Leucine-rich repeats, putative disease resistance protein and VQ proteins) with functions related to plant growth, development, and defense mechanisms. The marker-effect prediction was able to identify alleles responsible for predicting high yield and low Striga damage rating in the breeding panel. This study provides valuable insight for marker validation and deployment for Striga resistance breeding in maize.

Genomic selection has many applications within individual programs. Here, we discuss the benefits of forming a GS-based breeding consortium (GSC) among programs within the context of a recently formed a GSC of soft red winter wheat breeding programs. The GSC will genotype lines from each member breeding program (MBP) and conduct cooperative phenotyping. The primary GSC benefit is that each MBP can use GS to predict the local and broad value of all germplasm from all MBPs including lines in the early stages of testing, thus increasing the effective size of each MBP without significant new investment. We identified eight breeding aspects that are essential to GSC success and analyzed how our GSC fits those criteria. We identified a core of >5700 related lines from the MBPs that can serve in training populations. Germplasm from each MBP provided breeding value to other MBPs and program-specific adaption was low. GS accuracy was acceptable within programs but was low between programs when using training populations with little testing connectivity, but increased when using data from trials with high testing connectivity between MBPs. In response we initiated sparse-testing with a germplasm sharing scheme utilizing family relationship to connect our phenotyping of early-stage lines.

Iranian wheat landrace accessions (IWAs) were collected from country-wide farm fields and market places in 1935 by a professor at the University of Tehran and shared with University of California at Davis, California. IWAs were further submitted to the genebank of International Maize and Wheat Improvement Center (CIMMYT), Mexico. 2,403 IWAs from CIMMYT’s genebank were assayed by DArT-seq technology to assess genetic diversity. No apparent ecogeographic patterns related to genetic diversity were detected, probably due to long-term transport and frequent interchange of landraces among farmers. A multivariate clustering procedure combining genotypic and phenotypic information was used in selecting a core-set, which represented 15% of the hexaploid wheat accessions included in this study. This subset captured an estimated 93% of rare (frequency <0.05) alleles. Multisite phenotypic data (India, Mexico) validated the ability of the core-set in detecting useful variants. Potential donor accessions for multiple traits (disease resistance, zinc concentration) were identified from the core-set for wheat-breeding. This report illustrates a breeder friendly robust core-set formulation strategy for harnessing the useful genetic variation stored in the genebanks.

Septoria tritici blotch, caused by the fungus Zymoseptoria titici, poses serious and persistent challenges to wheat cultivation in Ethiopia and worldwide. Deploying resistant cultivars is a major component of controlling septoria tritici blotch (STB). Thus, the objective of this study was to elucidate the genomic architecture of STB resistance in an association panel of 178 bread wheat genotypes. The association panel was phenotyped for STB resistance, phenology, yield, and yield-related traits in three locations for 2 years. The panel was also genotyped for single nucleotide polymorphism (SNP) markers using the genotyping-by-sequencing (GBS) method, and a total of 7,776 polymorphic SNPs were used in the subsequent analyses. Marker-trait associations were also computed using a genome association and prediction integrated tool (GAPIT). The study then found that the broad-sense heritability for STB resistance ranged from 0.58 to 0.97 and 0.72 to 0.81 at the individual and across-environment levels, respectively, indicating the presence of STB resistance alleles in the association panel. Population structure and principal component analyses detected two sub-groups with greater degrees of admixture. A linkage disequilibrium (LD) analysis in 338,125 marker pairs also detected the existence of significant (p ≤ 0.01) linkage in 27.6% of the marker pairs. Specifically, in all chromosomes, the LD between SNPs declined within 2.26–105.62 Mbp, with an overall mean of 31.44 Mbp. Furthermore, the association analysis identified 53 loci that were significantly (false discovery rate, FDR, <0.05) associated with STB resistance, further pointing to 33 putative quantitative trait loci (QTLs). Most of these shared similar chromosomes with already published Septoria resistance genes, which were distributed across chromosomes 1B, 1D, 2A, 2B, 2D, 3A,3 B, 3D, 4A, 5A, 5B, 6A, 7A, 7B, and 7D. However, five of the putative QTLs identified on chromosomes 1A, 5D, and 6B appeared to be novel. Dissecting the detected loci on IWGSC RefSeq Annotation v2.1 revealed the existence of disease resistance-associated genes in the identified QTL regions that are involved in plant defense responses. These putative QTLs explained 2.7–13.2% of the total phenotypic variation. Seven of the QTLs (R2 = 2.7–10.8%) for STB resistance also co-localized with marker-trait associations (MTAs) for agronomic traits. Overall, this analysis reported on putative QTLs for adult plant resistance to STB and some important agronomic traits. The reported and novel QTLs have been identified previously, indicating the potential to improve STB resistance by pyramiding QTLs by marker-assisted selection.

Maintaining winter wheat (Triticum aestivum L.) productivity with more efficient nitrogen (N) management will enable growers to increase profitability and reduce the negative environmental impacts associated with nitrogen loss. Wheat breeders would therefore benefit greatly from the identification and application of genetic markers associated with nitrogen use efficiency (NUE). To investigate the genetics underlying N response, two bi-parental mapping populations were developed and grown in four site-seasons under low and high N rates. The populations were derived from a cross between previously identified high NUE parents (VA05W-151 and VA09W-52) and a shared common low NUE parent, 'Yorktown.' The Yorktown × VA05W-151 population was comprised of 136 recombinant inbred lines while the Yorktown × VA09W-52 population was comprised of 138 doubled haploids. Phenotypic data was collected on parental lines and their progeny for 11 N-related traits and genotypes were sequenced using a genotyping-by-sequencing platform to detect more than 3,100 high quality single nucleotide polymorphisms in each population. A total of 130 quantitative trait loci (QTL) were detected on 20 chromosomes, six of which were associated with NUE and N-related traits in multiple testing environments. Two of the six QTL for NUE were associated with known photoperiod (Ppd-D1 on chromosome 2D) and disease resistance (FHB-4A) genes, two were reported in previous investigations, and one QTL, QNue.151-1D, was novel. The NUE QTL on 1D, 6A, 7A, and 7D had LOD scores ranging from 2.63 to 8.33 and explained up to 18.1% of the phenotypic variation. The QTL identified in this study have potential for marker-assisted breeding for NUE traits in soft red winter wheat.

Understanding the mechanisms governing complex traits variation is a requirement for efficient crop improvement. In this study, the molecular characterization, marker-trait associations and the possibility for genomic selection in a collection of 281 Kersting's groundnut accessions were carried out. The diversity panel was phenotyped using an Alpha lattice design with two replicates in two contrasting environments. Accessions were genotyped using genotyping by sequencing technology. Genome-wide association analyses were performed between single nucleotide polymorphism markers and yield-related traits across tested environments. SNP markers were used to calculate the observed (Ho) and expected heterozygosity (He), and the total gene diversity (Ht). Genetic differentiation among accessions across ecological regions of origin was analysed. Our results revealed 493 quality SNPs of which 113 had a minor allele frequency>0.05, a total gene diversity of 0.43 and average Ho and He values of 0.04 and 0.22, respectively. Four clusters, highly differentiated by seed coat colour (Fst = 0.79), were identified. The population structure analysis showed two subpopulations with high differentiation across ecological regions (Fst = 0.37). The GWAS revealed 10 significant marker-trait associations, of which six SNPs were consistent across environments. The genomic selection through cross-validation showed moderate to high prediction accuracies for leaflet length, seed dimension traits, 100 seed weight, days to 50% flowering and days to maturity. This demonstrates the existence of genetic variability within Kersting's groundnut and shows the potential for the improvement of the species. The findings also provide a first insight into the phenotype-to-genotype relationships in Kersting's groundnut, using SNP markers.

In order to integrate genomics in breeding and development of drought-tolerant groundnut genotypes, identification of genomic regions/genetic markers for drought surrogate traits is essential. We used 3249 diversity array technology sequencing (DArTSeq) markers for a genetic analysis of 125 ICRISAT groundnut mini core collection evaluated in 2015 and 2017 for genome-wide marker-trait association for some physiological traits and to determine the magnitude of linkage disequilibrium (LD). Marker-trait association (MTA) analysis, probability values, and percent variation modelled by the markers were calculated using the GAPIT package via the KDCompute interface. The LD analysis showed that about 36% of loci pairs were in significant LD (p < 0.05 and r2 > 0.2) and 3.14% of the pairs were in complete LD. The MTAs studies revealed 20 significant MTAs (p < 0.001) with 11 markers. Four MTAs were identified for leaf area index, 13 for canopy temperature, one for chlorophyll content and two for normalized difference vegetation index. The markers explained 20.8% to 6.6% of the phenotypic variation observed. Most of the MTAs identified on the A subgenome were also identified on the respective homeologous chromosome on the B subgenome. This could be due to a common ancestor of the A and B genome which explains the linkage detected between markers lying on different chromosomes. The markers identified in this study can serve as useful genomic resources to initiate marker-assisted selection and trait introgression of groundnut for drought tolerance after further validation.

KEY MESSAGE:Genomic selection using data from an on-going breeding program can improve gain from selection, relative to phenotypic selection, by significantly increasing the number of lines that can be evaluated. The early stages of phenotyping involve few observations and can be quite inaccurate. Genomic selection (GS) could improve selection accuracy and alter resource allocation. Our objectives were (1) to compare the prediction accuracy of GS and phenotyping in stage-1 and stage-2 field evaluations and (2) to assess the value of stage-1 phenotyping for advancing lines to stage-2 testing. We built training populations from 1769 wheat breeding lines that were genotyped and phenotyped for yield, test weight, Fusarium head blight resistance, heading date, and height. The lines were in cohorts, and analyses were done by cohort. Phenotypes or GS estimated breeding values were used to determine the trait value of stage-1 lines, and these values were correlated with their phenotypes from stage-2 trials. This was repeated for stage-2 to stage-3 trials. The prediction accuracy of GS and phenotypes was similar to each other regardless of the amount (0, 50, 100%) of stage-1 data incorporated in the GS model. Ranking of stage-1 lines by GS predictions that used no stage-1 phenotypic data had marginally lower correspondence to stage-2 phenotypic rankings than rankings of stage-1 lines based on phenotypes. Stage-1 lines ranked high by GS had slightly inferior phenotypes in stage-2 trials than lines ranked high by phenotypes. Cost analysis indicated that replacing stage-1 phenotyping with GS would allow nearly three times more stage-1 candidates to be assessed and provide 0.84-2.23 times greater gain from selection. We conclude that GS can complement or replace phenotyping in early stages of phenotyping.