1. Molecular breeding project design and consultation
Molecular breeding is a process that applies molecular marker technology to conventional breeding, in order to develop better breeding materials more effectively in a shorter period. Molecular breeding technology has been incorporated into the whole process of conventional breeding (see picture below). Proper application of molecular breeding methods can increase selection intensity/scope and shorten the breeding cycle, hence develop more elite hybrids quicker. We thoroughly analyze the breeding process based on specific customer needs; provide a full package of molecular breeding strategy and execution plan.  

2. Deep understanding of breeding materials

The most direct and effective way to understand breeding materials is through multiple years and multiple locations’ field trial data, but this is greatly limited by labor and field resources. When introducing new breeding materials with unknown pedigree, molecular markers can be used to analyze large amount of materials very quickly. This provides analysis, diagnosis and understanding based on genetic characteristics. We analyze the breeding materials with high density SNP markers, provide a complete report with the information including but are not limited to: genetic diversity analysis, inbred lines comparison, heterotic group classification, and inbred test cross strategies with field trial information etc.

A. Genetic diversity analysis and inbred comparison

In the breeding process, only heritable genetic components will be inherited by the progenies. The elite parent lines with many good genes have higher probability to produce elite progeny lines. Therefore, it is very important to analyze and understand the genetic diversity in the germplasm. Genetic diversity is the overall measurement of genetic component difference in the germplasm. Broader genetic diversity means richer genetic resources in the breeding population, so there are more possible combinations to select and produce elite progenies.

Inbred lines are stable breeding materials and can pass genetic information down easily. Elite inbreds usually have large amount of field testing data. Combining with the genetic marker information, we can analyze and discover genetic segments that are associated with different traits, characterize disease resistances and abiotic stress resistance genes. Comparing new breeding materials such as hybrid derivatives with elite breeding materials through molecular markers, we can quickly understand genetic characteristics of new materials and design effective breeding strategies accordingly.

B. Definition of heterotic groups
Heterotic groups are generated naturally through long term reciprocal recurrent selection in corn. The paternal group and maternal group have strong heterosis in between, so there is bigger chance to create elite hybrids. Different heterotic group has enriched unique set of good genes, which can be detected through molecular markers. For those inbred lines with unclear heterotic classification or those newly introduced breeding materials, molecular markers can be used to define heterotic groups. In combination with field validation we can provide precise breeding strategies.

In the graph above, each line represents a unique corn variety. Red lines are from the heterotic group I, black lines are from the heterotic group II. Blues lines are the varieties which can hardly be grouped into any heterotic group. Longer distance between lines means larger genetic distance, which brings stronger potential heterosis between them.

C. Hybrid performance prediction
Accurate classification of heterotic group is very important for inbred development and hybrid creation. When creating hybrid varieties, parental lines must be chosen from different heterotic groups. For example the two parental lines of XY335 are from SS and NSS heterotic groups. On the other hand, the most effective way for inbred development is to breed within the heterotic group.

As shown below, in a DH population derived from F2, only a small portion of the progenies at both ends of the distribution belong to opposite heterotic groups. There is big genetic difference, so it is more likely that elite hybrids can be created between them. Using molecular markers, we can detect paternal-alike and maternal alike progenies, avoiding the mixtures.

3. Hybrids prediction
In conventional breeding, the breeding value for the inbred and hybrid varieties can only be estimated through field testing, which is very limited by labor and field resources. And complex traits are hard to observe and test given strict environment requirement. Utilizing modern genetics methods and bio-statistics models, we can estimate breeding values for existing hybrids,predict unobserved traits or untested hybrid combinations, as long as thereare field trial information and molecular marker data.   

A. Breeding value calculation
Breeding value is the summation of gene average effects in a variety. Variety with higher breeding value carries more elite genes and has higher chance to generate elite progenies. Using molecular marker information, we can use BLUP (Best Linear Unbiased Prediction) method to calculate GCA(General Combining Ability), SCA(Specific Combining Ability). This is called G-BLUP (Genomic-BLUP) and can be done even without the pedigree information. 

B. Hybrids prediction
Associating phenotypic and molecular marker data, we use mixed linear model to predict phenotypic values for different traits. Higher correlation between observed and predicted values means higher predictive power in general. For traits and breeding materials with high predictive power, we can predict their progeny’s field performance without field trial. This can help to minimize field work and shorten the breeding cycle.

C. Hybrid creation recommendation
For a group of key inbred lines, we design an efficient way to create and test a small subset of all possible hybrid combinations. Using the field trial data from the subset, in combination of key inbreds’ marker data, we predict field performance of all the possible hybrid combinations. We recommend good hybrid combinations for further testing, which can save you a lot of time and effort in hybrid creation and only focus on hybrids with high potential.

4. Marker-assisted selection
A. Marker-assisted selection
Marker-Assisted Selection (MAS) can quickly analyze genetic components in each breeding line and conduct direct selection at genotypic level. We use SNP markers closely linked with traits to screen large amount of seedlings or seeds, and make selection decision at early stage. This speeds up selection process and saves field trial cost.

The traits that we can conduct Marker-Assisted Selection include: 
Corn Disease:
Heat Smut, Stalk Rot, Northern Leaf Blight, Southern Leaf Blight, Dwarf Mosaic Virus Disease, Maize Rough Dwarf Disease, Grey Leaf Spot. 
Stress Resistance: Nitrogen/Phosphate Use Efficiency, Salt Tolerance.
Other Corn Traits: Flowering Date Improvement, Seed Oil Content improvement.
Wheat Disease: Stripe Rust, Fusarium Head Blight.

Allele a is susceptible allele of a marker closely linked to Fusarium Moniliforme Ear Rot and allele B is the resistant allele. Therefore we can use this marker to select the segregating progenies. Progeny 1 and 4 have allele b, but progeny 2 and 3 have Allele a of the marker. We can predict that progeny 1 and 4 will have better resistance to Fusarium Moniliforme Ear Rot disease, and progeny 2 and 3 are more likely to be susceptible to the disease.

B. Marker-assisted backcrossing
Backcrossing is the most effective way to introduce trait associated gene(s) from a donor parent into a recurrent parent. It is broadly used in inbred improvement. Using molecular markers, we can not only select for donor segment(s), but also select on recurrent parent genetic background. This makes the whole process more accurate and quicker. We provide services on backcrossing project design and consultation, population creation recommendation, molecular marker detection, and backcrossing single plant selection recommendation etc.

Here is a comparison between conventional back crossing (Left side) and marker assisted back crossing (Right side) experiment process. Marker assisted back crossing can recover the recurrent parent genome to more than 99% in three generations. This saves at least three generations in comparison to conventional method. On the other hand, the selection on target genes by Marker-Assisted Selection can guarantee not losing the target in the back crossing process.