thumbnail_small_OGMQXH_0_0072d6e25f_f62446c3be.png

webp_small_OGMQXH_0_0072d6e25f_f62446c3be.webp

We regularly release blogs written by statisticians, researchers, scientists and industry leaders. From how-to guides to the application of data analytics and opinion pieces, you will find a wealth of content below

Welcome to VSNi Blogs

Trusted and powerful data analysis tools for your discovery needs

Uncover our statistics software

thumbnail_Linkedin Content (43).png

webp_Linkedin Content (43).webp

The Power of GWAS with ASReml-R

Overcoming the Complexity of Modelling Spatial Variation 

COMPETITIVE GENETICS: EXPLORING THE IMPACT OF DIRECT AND INDIRECT GENETIC EFFECTS IN TREE BREEDING



thumbnail_Linkedin Content (30).png

webp_Linkedin Content (30).webp

Competitive Genetics: Exploring the impact of direct and indirect genetic effects in tree breeding

A Journey with Genstat: A Conversation with Mewa S. Dhanoa 

thumbnail_Linkedin Content (29).png

webp_Linkedin Content (29).webp

Data, data everywhere…but is it helping your analytics?

thumbnail_November - December landscape (8).png

webp_November - December landscape (8).webp

A Snippet for Error Handling in R: An Example with ASReml-R

thumbnail_November - December landscape (6).png

webp_November - December landscape (6).webp

Genetic correlations, what are they, and why do we need them? 

thumbnail_Linkedin Content (16).png

webp_Linkedin Content (16).webp

Overlaying Design Matrices for a Parental Model in ASReml-R

Data analysis is essential for addressing the challenges of food production and we at VSNi envision a future where these challenges are eradicated through the power of extracting meaningful insights from data.

VSNi's vision: A world without hunger, powered by data-driven breeding 

Why put your trust in ASReml? In this blog we explain why it is the go-to tool for scientists and researchers

ASReml: The powerhouse statistical software tool where speed meets accuracy 

Statistical inference is the process of drawing some conclusions about a population based on the sample data at hand.

thumbnail_medium_TWINIQ_2_Header_546db569dd_54daa15b6c.jpeg

webp_medium_TWINIQ_2_Header_546db569dd_54daa15b6c.webp

Likelihood-based statistical inference for variance components in linear mixed models

An important aim when fitting linear mixed models (LMM) is the use of random effect estimates. In some analyses, such as genetic evaluations, the main objective of the analysis is to obtain these estimates.

Calculating accuracy and reliability of random effect estimates with ASReml-R

The programme is meticulously designed to equip you with the expertise to use ASReml-R version 4.2 for the analysis of genetic data. ASReml-R is a statistical software (and library in R) that fits linear mixed models using REML methodology and calculates BLUP values. 

Genetic Analyses with ASReml-R training course 

Latin Squares are a very well-known trial design, second only to Randomized Blocks. They are formed from a square layout of plots, with the same number of replicates as there are treatments. 

A little bit of Latin: Latin squares and Latinized designs

Some statistical practitioners routinely fit GLMMs using the HGLM framework. In this blog we discuss why. 

thumbnail_blog header glmm as hglm.webp

webp_blog header glmm as hglm.webp

Why fit a GLMM as an HGLM?

For a farmer, the disease raises crucial questions: How quickly will symptoms develop, and will welfare and productivity suffer? 

thumbnail_sheep with MV (1).webp

webp_sheep with MV (1).webp

Uncovering disease control strategies: A Genstat case study on Maedi-Visna (MV) epidemiology in sheep 

Maximizing Field Trial Data for Agricultural Advancements 

thumbnail_Linkedin Content (55) (1).webp

webp_Linkedin Content (55) (1).webp

For simple linear regression with groups, why do we first allow for different intercepts and then for different slopes?

thumbnail_blog_header_regression_with_groups_panel.webp

webp_blog_header_regression_with_groups_panel.webp

Regression with groups: 
Testing for group-specific intercepts and slopes


Learn how to efficiently identify various taxa in fields like medicine, agriculture, and hobbies. Explore the concept of fixed results and exact identification, where specific attributes reliably define taxa.

thumbnail_blog_header_identification_methods.webp

webp_blog_header_identification_methods.webp

Efficient identification methods: Constructing irredundant test sets for confident taxa identification

ASReml-R is sophisticated statistical software designed specifically for the analysis of linear mixed models, making it ideal for use in animal, aquaculture and plant breeding

thumbnail_blog_header_asreml-r_4.2.webp

webp_blog_header_asreml-r_4,2.webp

ASReml-R version 4.2: More power and speed for breeding analyses

This blog discusses a-posteriori techniques that can be used for improving the quality of the information extracted from an already established and measured field, laboratory or greenhouse experiment. 

thumbnail_blog header post hoc blocking.webp

webp_blog header post hoc blocking.webp

Improving the design of your experiments part 2: post-hoc blocking

Precision breeding is poised to become one of the most influential genetic improvement tools.

thumbnail_blog header precision breeding.webp

webp_blog header precision breeding.webp

Elevating animal breeding and aquaculture: embracing data-driven precision to meet global challenges

If you are a Python user you might have noticed that despite myriad available libraries and frameworks - especially on the Machine Learning side - this language still lacks depth on some conventional, well-established and proven statistical tools, such as linear mixed models (LMM).

thumbnail_blog_header_python_to_R.webp

webp_blog_header_python_to_R.webp

Fitting mixed models in Python with ASReml-R

For incomplete block designs, incorporating some good a-priori blocking at the stage of planning is always a good practice

thumbnail_blog header a-priori blocking.webp

webp_blog header a-priori blocking.webp

Improving the design of your experiments part 1: a-priori blocking

The 23rd edition of Genstat includes many more useful extensions and new features to help you perform your desired statistical task. 

thumbnail_blog header Gen 23.png

webp_blog header Gen 23.webp

Genstat 23 out now

Multi-environmental trials (MET) are an essential part of plant breeding programmes to select the best cultivars for a zone or for specific environments

thumbnail_blog_header_managing_datasets_MET.jpeg

webp_blog_header_managing_datasets_MET.webp

Managing datasets for multi-environmental trials: best practices 

Unravelling natures blueprint: Understanding Heritability and genetic influence in plant and animal breeding

thumbnail_9.png

webp_9.webp

Are blocks fixed or random?

VSNi: Empowering Researchers with Innovative Tools

VSNi: Empowering researchers with innovative tools

In this blog we'll explore the relationships between the linear model, GLM, GLMM and HGLM models

Extending linear models to accommodate non-Normal data and random effects: 
LM → GLM → GLMM → HGLM


Examples of using the exponential function

The exponential function: a non-linear regression that may be useful to biologists - part 2

Simple linear regression is familiar to almost everyone: a straight line fitted through points displayed on an “X-Y” graph

The exponential function: a non-linear regression that may be useful to biologists - part 1

In Bayesian analysis, A is a parameter or parameters we wish to estimate, and B is a set of data we have collected.

Bayesian analysis: What is it?

thumbnail_Copy of September.png

webp_Copy of September.webp

Is envirotyping beneficial to my breeding program? 

Pearson correlation is frequently used to measure the strength and direction of the linear association between two numeric variables

thumbnail_Linkedin Content (31).png

webp_Linkedin Content (31).webp

Genstat: The ideal tool for plant breeding trial analysis 

thumbnail_Header.jpg

webp_Header.webp

Moving from confidence in 1-D to 2-D

Principal Components Analysis (PCA), a popular multivariate method, aims to reduce the dimensionality of multivariate data

thumbnail_blog-header-PCA-patterns.jpg

webp_blog-header-PCA-patterns.webp

PCA: Hunting for patterns and structure in my data

To ensure the validity of an analysis of variance (ANOVA) some important assumptions are made, including that the residuals are Normally distributed and with constant variance

thumbnail_blog-header-anova-residual-diagnostic-plots.jpg

webp_blog-header-anova-residual-diagnostic-plots.webp

Checking those all-important ANOVA residual diagnostic plots

Transforming an estimated mean back to the original scale is fairly straightforward: we just apply the inverse function of the transformation.

thumbnail_blog-header-back-transform.jpg

webp_blog-header-back-transform.webp

Can you back-transform a standard error?

A separation plot is a graphical approach for assessing the fit of a model with a dichotomous (i.e., binary) or polytomous (i.e., multi-categorical) outcome

thumbnail_separation plots 10.jpg

webp_separation plots 10.webp

Separation plots – an extremely useful tool for visualising the fit of models with categorical outcomes

Unreplicated trials lack the accuracy we need to select the outstanding genotypes, but this does not mean they are not useful! 

thumbnail_blog_header_unreplicated-trials-part2.jpg

webp_blog_header_unreplicated-trials-part2.webp

Unreplicated trials: Can they provide the accuracy we need? Part 2

A commonly made mistake when interpreting p-values is to accept the null hypothesis when the resulting p-value isn’t deemed small enough. 

thumbnail_blog header equivalence non-inferiority.jpg

webp_blog header equivalence non-inferiority.webp

Neither better nor worse? Equivalence, non-inferiority and non-superiority tests

In scientific papers and reports, experimental designs are often described as “resolvable” or “non-resolvable”. But what does this mean and why does this matter? 

thumbnail_blog header resolvable designs.jpg

webp_blog header resolvable designs.webp

Resolvable or non-resolvable design:
What does it mean and why does it matter?


Is the use of LD panels really the best use of your monetary resources? Is the use of LD panels a good long-term strategy? 

thumbnail_blog header LD chips.jpg

webp_blog header LD chips.webp

The long-term implications of Low-density SNP panels

thumbnail_outlier protest.png

webp_outlier protest.webp

Christmas crackers: VSNi's favourite statistical jokes

In GWAS, the whole genome is studied for genomic variants associated with a particular trait. 

thumbnail_manhattan plot.jpg

webp_manhattan plot.webp

An overview of genome-wide association studies (GWAS)

Mistakes unfortunately happen in experimental design, statistical analysis and interpretation of statistical results

medium_blog header spot error.jpg

small_blog header spot error.jpg

thumbnail_blog header spot error.jpg

Can you spot the statistical flaws?

The statistical theory of estimating the population parameter proportion is relatively well understood

small_blog estimating proportions header.jpeg

thumbnail_blog estimating proportions header.jpeg

Critical considerations for estimating proportions

The following strategy should lead to a sensible model for data from a designed experiment

small_blog header strategy.jpg

thumbnail_blog header strategy.jpg

Strategy for a linear mixed model analysis of a designed experiment

conditional residuals are the difference between the observed data and the conditional fitted values.

small_blog header goodness of fit.jpg

thumbnail_blog header goodness of fit.jpg

A lightning introduction to linear mixed models for designed experiments: part 4

The estimates of the fixed effects used with REML are often called BLUEs, which is an acronym for Best Linear Unbiased Estimates. 

small_blog header the BLUEs.jpg

thumbnail_blog header the BLUEs.jpg

A lightning introduction to linear mixed models for designed experiments: part 3

For the majority of statistical analyses, we find ourselves with the ‘Multiple Comparison Problem’.

small_blog--header-multiple-comparison-problem.jpg

thumbnail_blog--header-multiple-comparison-problem.jpg

The Multiple Comparison Problem

The problem with adding more data to achieve statistical significance is that in doing so you increase the probability of obtaining a significant effect by random chance.

small_blog_header_dont_add_data.jpg

thumbnail_blog_header_dont_add_data.jpg

Don’t add more data to make your results significant

The small ruminant lentiviruses (SRLV, family retroviridae) are recognised as a heterogeneous group of viruses that infect sheep, goats and wild ruminants

small_sheep.jpeg

thumbnail_sheep.jpeg

Quantification holds the key to controlling disease

All breeding programs focus on selecting individuals with outstanding breeding values (BVs). But the concept of BVs is sometimes hard to grasp

Why does a breeding value depend on other breeding values?

Many analysts are now using linear mixed models (LMMs), also known as multi-level models, to analyse data from designed experiments instead of ANOVA. So, which method should you use?

small_blog_anova_or_LMM_header.jpg

thumbnail_blog_anova_or_LMM_header.jpg

ANOVA or linear mixed model?

In this blog we'll give a brief overview of some of the new features and enhancements in Genstat 22nd edition

small_Genstat splash screen.jpg

thumbnail_Genstat splash screen.jpg

New features and enhancements in Genstat 22nd edition

In this blog, we will illustrate replication, randomization and blocking using a specific example in order to give you a more intuitive understanding of their importance.

medium_random numbers.jpg

small_random numbers.jpg

thumbnail_random numbers.jpg

This randomization is not for you!

Breeders investigate the G×E interaction to identify genotypes well suited for use in many environments, as well as those with exceptional performance in specific environments

Software

Services

Sector

Learning

Resources

Blogs

About us

Licensing

Contact

Welcome to VSNi Blogs

We regularly release blogs written by statisticians, researchers, scientists and industry leaders. From how-to guides to the application of data analytics and opinion pieces, you will find a wealth of content below

Home

Blogs

Most Recent

READ MORE

The Power of GWAS with ASReml-R

READ MORE

Overcoming the Complexity of Modelling Spatial Variation

READ MORE

Competitive Genetics: Exploring the impact of direct and indirect genetic effects in tree breeding

READ MORE

A Journey with Genstat: A Conversation with Mewa S. Dhanoa

READ MORE

Data, data everywhere…but is it helping your analytics?

READ MORE

A Snippet for Error Handling in R: An Example with ASReml-R

READ MORE

Genetic correlations, what are they, and why do we need them?

Popular

Competitive Genetics: Exploring the impact of direct and indirect genetic effects in tree breeding

22 February 2024

Linear mixed models: a balanced lattice square

17 February 2021

Predicting the Performance of Genotypes by Multi-Environment Trial Analyses

21 January 2021