SHARE STOP SHARE   Compare Projects



New Project Name:

Attention, there are {{warnings.length}} warnings! R Simulation most likely cannot be run unless they are fixed.


Warnings

Warning {{index}} : {{war}}






General Info

{{key}} : {{value}}
Node Info

{{key}} : {{value}}
Edge Info

{{key}} : {{value}} ({{individualsVar_options[active_edge.useVar.indexOf(key)].name}})

MoBPS

MoBPS    {{geninfo['Project Name']}}

{{geninfo['sharedWith'] }}

Load a new/existent project from your own database:

Welcome to MoBPS. Please create a project!

Exemplary Templates:
   Version:

-- If you are a new MoBPS user, there will not be any projects to display in the grouping projects section.
    • You could scroll down to General Information section to enter the columns and Save.
-- If you are an existing user and already have projects on your own, All existing projects would be under the Common_Projects folder.
    • You can create a folder and group your projects
-- Edit Group : Click Edit and change your own group name and enter to update.
-- Remove Group: Click Remove to remove projects under this folder.
-- Add Sub Folder: Click Add Sub Folder, the sub-folder will be added just below to the group.
    • You can drag and drop projects to the newly created folder (or)
    • you can drag the folder to keep it as a main group too!
-- Edit/display the project - Click your project from your folder then the project is in Edit mode.

You are assigned to User Class Admin. This enables you to use 20 Cores, 40 GB Max-Memory and maximum run time of 120 hours

You are assigned to User Class Professional. This enables you to use 5 Cores, 30 GB Max-Memory and maximum run time of 48 hours

You are assigned to User Class Student. This enables you to use 2 Cores, 20 GB Max-Memory and maximum run time of 4 hour

You are assigned to User Class Test. It is not possible to use our backend-server for simulation. To simulate download the json-file run via json.simulation() in R.

• Enter your breeding program in the following modules.
• ⓘ buttons indicate additional information when hovering over.
• Use the navigation bar to jump between modules, change your password, etc.
• In case our server resources are not enough for you please press the Export button
• You can run simulation on your own device in R via: json.simulation("FILENAME.json")
• Script with uploaded genotype / maps will only run on our server!
• We are happy to help you set up your own MoBPS web-server!
• Basic tutorials are in preparation and can be found at: https://www.youtube.com/channel/UC4LDcBka39NidOF1y_65FFw
• For advanced questions, bug reports, potentially collaborations contact me! (Torsten Pook / torsten.pook@uni-goettingen.de)

General Information

--In this module you can enter basic information on the species you are working on
--You can select from provided maps (via Ensembl / MoBPSmaps) or upload you own .map-file:
--Columns for a customize upload are: Chromosome, SNP-name, physical position in base-pairs, position in Morgan (if NA -> bp / 100.000.000), allele frequency (optional)
--Most additional modules on default are hidden! Press Advanced settings to unlock more complicated options

Project Name ⓘ Enter the name of your project. You can save/copy/delete your project via the action bar and load different version to return to.
Advanced settings
Test-Mode / Size Scaling ⓘ Activating this will unlock the scaling of node sizes and ignore imported genotypes.
Advanced Trait modelling ⓘ Activating this will unlock use of dominant and epistatic trait architectures.
Non-additive effects ⓘ Activating this will enable generation of non-additive QTL effect
Repeatability ⓘ Activating this will enable to set a repeatability unequal to heritability
Maternal / paternal effects ⓘ Activating this will enable modelling of maternal / paternal effects
Traits as combination of other traits ⓘ Activating this will enable modelling features so traits are modelled as linear combinations of other traits
Transformation function ⓘ Activating this will enable application of transformation functions on trait phenotypes
Trait rescaling ⓘ Activating this will enable rescaling of traits after a couple of simulated generations
LD build-up Module ⓘ Activating this will unlock a module to simulated genomic data with baseline LD as a starting point of the simulation
Culling Module ⓘ Activating this will unlock the culling module to simulate the death of individuals. Death individuals can still be used for BVE but not for Reproduction
Subpopulation Module ⓘ Activating this will unlock the specification of subpopulation with different allele frequencies and trait architectures.
Economic Module ⓘ Activating this will unlock the use of economic parameters.
Population History ⓘ Activating provide you with additional features for the simulation of thousands of generations
Litter size Module ⓘ Activating provide you with additional features to control the litter size
Modify multiple nodes/edges ⓘ Activating provide you with additional features to modifiy multiple nodes / edges simultaneously
Advanced Edge/Node options ⓘ Activating this will unlock additional options for nodes/edges
Share genotyped / Multiple Arrays ⓘ Activating this will unlock additional options for nodes/edges
Max offspring ⓘ Activating this will unlock additional options for nodes/edges
Avoid Half/Fullsib matings ⓘ Activating this will unlock additional options for nodes/edges
OGC ⓘ Activating this will unlock additional options for nodes/edges
Selection ratio ⓘ Activating this will unlock additional options for nodes/edges
Threshold selection ⓘ Activating this will unlock additional options for nodes/edges
Advanced input phenotype ⓘ Activating this will unlock additional options for nodes/edges
Skip BVE ⓘ Activating this will unlock additional options for nodes/edges
Calculate reliability ⓘ Activating this will unlock additional options for nodes/edges
Use last available ⓘ Activating this will unlock additional options for nodes/edges
Delete data ⓘ Activating this will unlock additional options for nodes/edges
Ignore Size scaling ⓘ Activating this will unlock additional options for nodes/edges
Copy settings from other nodes/edges ⓘ Activating this will unlock additional options for nodes/edges
miraculix-active ⓘ Deactivating this will increase run-times but lead to RData population that can be analyzed without miraculix/RandomFieldsUtils
Parallel Computing + Multiple Simulation ⓘ Activating this will allow to run a simulation multiple times.
Export/Import Box ⓘ Activating provide you with a text box for import/export
Species ⓘ Select the species. This will unlock the correct Ensembl Datasets etc. Use a template to start with already entered somewhat realistic values.
Time Unit ⓘ Select the time unit used. Time of generation for each cohort will be displayed.
Mutation rate ⓘ Mutation Rate per meiosis when performing reproduction.
Genetic Data ⓘ Select the genome you want to work with.

Ensembl Dataset ⓘ Currently only the genomic maps available in MoBPSmaps are provided. Alternatively upload your own map or provide us with the map you want us to add in (torsten.pook@uni-goettingen.de). Please choose a species first.
Max. Number of SNPs ⓘ Only a subsample of the Ensembl dataset is used to save computing time. If nothing / non-numeric value is entered use the full dataset.
Upload Map File ⓘ Currently only .vcf .map and .RData-files are supported. Contact Torsten (torsten.pook (at) uni-goettingen.de) to add additional maps.
Own Map Path {{geninfo["Own Map Path"]}}
Number of Chromosomes
Chromosomes of Equal Length


Chromosome Physical Length (in MbP) ⓘ The human genome has a size of 3200 Mb. Most livestock/crops are smaller! Marker Density (SNP/Mbp) ⓘ Do not overdo it with markers! They oftentimes mostly add memory requirement and computation time. Recombination (cM/Mbp) ⓘ 1cM/Mbp is commonly assumed. E.g. chicken have higher recombination rates (~3-4 cM/Mbp).
Chromosome 1 - {{geninfo['Number of Chromosomes']}}


Chromosome Physical Length (in MbP) Marker Density (SNP/Mbp) Recombination (cM/Mbp)
Chrom {{chromo_display.indexOf(chrom)+1}}
Advanced Test-mode:
Test/Speed Mode ⓘ Activate this to set the number of repeats to 1, reduce the genome length to 10M and 200 SNPs and deactivate data import. This should heavily reduce computation times for testing.
Individual numbers scaling ⓘ Scale the number of individuals in each node by this factor. This can massively reduce computing times! Point numbers are rounded.
Advanced parallel computing:
Number of Simulations ⓘ Number of simulations of the breeding program you want to perform. This will massively increase run-time but also reduce variance of results.
Number of Simulations run in parallel ⓘ Number of simulations of the breeding program to run simultaneously. If taking more cores than allowed this will automatically be reduced internally.
Number of cores used per simulation ⓘ Number of cores used per simulation. If taking more cores than allowed this will automatically be reduced internally.
Advanced population history:
Set a new base-line population every X - generations ⓘ Genotypes for a baseline/founder population are stored leading to less points of recombination/mutation.
Delete generations before base-line ⓘ Genotypes for a baseline/founder population are stored leading to less points of recombination/mutation.
Advanced trait rescaling:
Time point to apply trait rescaling ⓘ Semicolon separated list of time point to apply rescaling of trait mean/variance (e.g. 50;100)


Phenotype Information ⓘ Here you can design the traits you want to work with.

Press "Add new phenotype" to add additional traits
• Polygenic Loci are purely additive QTL but on a randomly selected SNP of the used array
• To model more complex traits activate the "Advanced Trait modelling" in the General Information
• For correlated traits the number of QTLs for each trait will be higher than entered as QTLs from other traits will also impact a given trait. Unless correlations are high these should only be small effect (see MoBPS Guidelines for details on the method)
Make sure you correlation matrices are positive definit - otherwise we will make them positive definit automatically!


Phenotype ⓘ Name of the trait. Unit ⓘ Optional. Unit of measurement for the trait. Pheno. Mean ⓘ The avg. Founder individual will have this mean. To have subpopulations with different means use the Subpopulation module. Pheno. SD ⓘ Trait variance for the group of all founders will be scaled accordingly. Residual variance is assumed to be fixed over all generations. Heritability ⓘ Heritability of the trait. Heritability ⓘ Heritability of the trait. Repeatability ⓘ Repeatability of the trait (If empty will be equal to heritability - independent observations). # Polygenic Loci ⓘ Number of QTL. To simulate non-additive QTL effect activate the advanced setting Complex Trait architecture. # additive QTL ⓘ Number of QTL. To simulate non-additive QTL effect activate the advanced setting Complex Trait architecture. # dominant QTL ⓘ Number of purely dominant QTL. # qualitative epistatic QTL ⓘ Number of QTL with qualitative epistatic effect between two markers. # quantitative epistatic QTL ⓘ Number of QTL with quantitative epistatic effect between two markers. # additive equal size QTL ⓘ Number of purely additive QTL with all QTLS having the same effect size. # dominant equal size QTL ⓘ Number of purely dominant QTL with all QTLS having the same effect size. # only positive dominant effects ⓘ Set this to make sure the heterozygous variant in each dominant QTL has the same as the effect as the better homozygous variant (e.g. for hybrid breeding). maternal trait ⓘ Genomic value of the trait is caused by the mother paternal trait ⓘ Genomic value of the trait is caused by the father Major QTL ⓘ Use this if you want to manually add single marker effects to SNPs with known effects. combination of traits ⓘ This trait is a linear combination of other traits combination of traits ⓘ This trait is a linear combination of other traits Value per unit (€) ⓘ Economic gain by increased unit of the trait. Only relevant for economic calculations. Currently only linear effect and no interactions - Contact Torsten if you need more! Value per unit (€) ⓘ Economic gain by increased unit of the trait. Only relevant for economic calculations. Currently only linear effect and no interactions - Contact Torsten if you need more! Value per unit (€) ⓘ Economic gain by increased unit of the trait. Only relevant for economic calculations. Currently only linear effect and no interactions - Contact Torsten if you need more! Apply Transformation ⓘ The only allowed input variable is the genomic value (x). Potential input: function(x){y = x^2; return(y)} Show Cor


Major QTL syntax ⓘ Select in which way to want to enter information on the position of the major QTLs.


Residual Correlation ⓘ Make sure this matrix is positive definit. Otherwise we will project the matrix to the space of positive definit matrices.
{{traitsinfo[rind]['Trait Name']}}
{{traitsinfo[rind]['Trait Name']}}

Enter Phenotypic correlation instead of residual correlation
Genetic Correlation ⓘ Make sure this matrix is positive definit. Otherwise we will project it so the space of positive definit matrices
{{traitsinfo[rind]['Trait Name']}}
{{traitsinfo[rind]['Trait Name']}}


Creating own Selection Indexes

Create your own selection indexes.
• Selection Indexes can be used in all edges in which individuals are selected.
• Values from which to select later (usually breeding values / phenotypes) are scaled before applying the selection index.
• Scaling can be performed per phenotypic, genomic or breeding value variance of the group of individuals to select from!


Creating own Phenotyping Classes

Create your own phenotyping classes
• Only numeric values are allowed with number indicating the number of observations collected.
• Unless repeatability is entered in the trait generation, observations are simulated as INDEPENDENT observations.
• Phenotyping classes are attributes that can be assigned to nodes in the breeding scheme.


Genotyping Arrays ⓘ

• On default all genotyping will use all available simulated markers
• Add additional arrays here
• You can selected which array to use in the node of the respective cohort

Array Name Number of SNPs ⓘ Allele frequencies are drawn from a Beta (p,q) - distribution
{{vv.Name}}

LD build-up ⓘ

• Founder genotypes on default have no LD / haplotype structure.
• This module provides options to perform generations of random mating to get a baseline population structure
• For more complex population history (e.g. bottlenecks, systematic drift etc) you need to simulate this via additional nodes in the breeding scheme
• If different founder nodes should have different population history add multiple Subpopulations in the Subpopulation Module
• This will alter the allele frequency spectrum selected in the Subpopulation module as it will lead to drift and thus usually include more rare / fixated markers

Subpopulation Individuals ⓘ Number of individuals to generate per generation Share female ⓘ Share of female individuals in each generation Generations ⓘ Number of generations to simulate. Larger populations usually require higher number of generations to build up stronger LD. Fix Major QTL frequency ⓘ Due to drift the allelel frequency of major QTLs can change in the LD build-up. By activating this, this will be negated. LD build-up will be negated for Major QTLs!
{{vv.Name}}

Culling ⓘ

• Define reasons for individuals to exit the breeding program for reproduction.
• These individuals can still be used in downstream BVEs
• Individuals with BV1 (in the index selected) will die with probability 'Share death at BV1'.
• In case of no underlying genomics all individuals will die with this probability.
• In case of a genetic reason provide BV2. Values for each genomic value will be derived based on linear extension.

Create your own culling reasons.
• Parameters are chosen according the culling module in breeding.diploid().
• Death individuals can still be used for BVE but not for reproduction.
• Individuals with BV1 (in the index selected) will die with probability 'Share death at BV1'.
• In case of no underlying genomics all individuals will die with this probability
• In case of a genetic reason provide BV2. Values for each genomic value will be derived based on linear extension.

Culling reason At age (in time unit) ⓘ At what age does the culling potentially occur. Usage for reproduction at the time point itself is still possible. Relevant for sex ⓘ Is this relevant for Male/Female/Both? Genetic Index for survival ⓘ Select a selection index linked to the culling Share death at BV1 ⓘ What share of individuals are culled when they have BV1 (genomic value * selected index). BV1 Share death at BV2 ⓘ What share of individuals are culled when they have BV2 (genomic value * selected index). BV2
{{vv.Name}}

Multiple Subpopulations

Create your own subpopulation.
• Each founder node of the breeding scheme can be assign to a subpopulation.
• Allele frequencies will be calculated based on all cohorts of the given subpopulation with imported genotypes.
• Otherwise: Each subpopulation has independently sampled allele frequencies from a Beta(p,q)-distribution.
• To model different level of diversity controll the share of fixated markers.
• In case of different mean trait values additional QTL effects are put on fixated markers.

Subpopulation Sampled allele frequency (p) ⓘ Allele frequencies are drawn from a Beta (p,q) - distribution Sampled allele frequency (q) ⓘ Allele frequencies are drawn from a Beta (p,q) - distribution Share of fixated markers in A ⓘ Define allele frequencies for multiple subpopulations. Allele frequencies are drawn from a Beta(p,q) distribution. To model different levels of diversity use Share of fixated markers. A population with less diversity should have more fixated markers. More similar population should be fixated in the same direction for often. This cannot replace the use of REAL genomic data or simulated data with population structure (this module is mostly intended for baseline testing). Share of fixated markers in B ⓘ Define allele frequencies for multiple subpopulations. Allele frequencies are drawn from a Beta(p,q) distribution. To model different levels of diversity use Share of fixated markers. A population with less diversity should have more fixated markers. More similar population should be fixated in the same direction for often. This cannot replace the use of REAL genomic data or simulated data with population structure (this module is mostly intended for baseline testing). Nr. of Markers with manual chosen allele frequency Deviation from Mean for Trait {{trait['Trait Name']}}
{{vv.Name}}

Litter size

• The entered litter size will be applied on all newly generated individuals.
• This does NOT include new cohorts generated via selection/combine/aging etc.
• In case probabilites do not add up to 1 this will automatically be scaled accordingly.

Litter size Probability


General Economy Parameters

• Assign costs to basic breeding actions.
• Fixed costs are assumed to happen at time 0.
• Interest is applied on all cost types.
• Housing classes can be selected for all nodes of the breeding scheme.

Fixed cost (€) ⓘ Fixed costs of the breeding program
Interest rate (%) ⓘ Applied interest rate for all costs occurring.
Genotyping cost per individual (€) ⓘ Genotyping costs per individual. Currently only genotyping with a single array is supported. Modes with different chips and marker densities are coming.
Housing/Field cost class ⓘ Type a name and then click on "Creating new HFC Class" to create your own classes of housing/field costs. These classes can then be assigned to each cohort in the breeding scheme. Housing/Field cost (€)
{{vv.Name}}

Declaring own variables

In this module you can assign numeric values to variables
Variables can as inputs for nodes and edges
The main intended use is when having a high number of nodes/edges with the same number (e.g. number of individuals) and simultaneously wanted to change values for all nodes/edges between simulation runs


Variable Name Variable Value
{{vv.name}}

Modify Multiple Nodes/Edges ⓘ Via this module multiple nodes/edges with similar attributed can be changed simultaneously.
Select which nodes/edges to change by setting requirements.


Change Requirements Parameter to change // Change to
Breeding Type

Sex - Parental Node

Sex - Child Node
Sex

Number of Individuals
(leave empty to select all)

Phenotyping Class

Housing Cost Class
Breeding Type
Time needed (leave empty to not change) ⓘ Provide the time this breeding action takes. Reproduction currently needs at least one time unit. {{geninfo['Time Unit']}}
Use last available ⓘ Activating this will use the last repeat available from that node - not necessary the node from the same repeat
OGC ⓘ Use optimum genetic contribution according to Wellmann et al 2019 to determine how often each individual is used for reproduction.
Avoid full-sib matings ⓘ Activate this to not use full-sibling pairs for reproduction.
Avoid half-sib matings ⓘ Activate this to not use half-sibling pairs for reproduction.
Maximal # offspring (leave empty to not change) ⓘ Maximum number of offspring each individual from the parent node is allowed to have.
Maximal # offspring from one mating pair (leave empty to not change) ⓘ Maximum number of offspring from one mating pair.
Selection ratio (leave empty to not change) ⓘ Use this if you want to use some individuals more frequently for reproduction. A value of 2 will lead to the individual will the highest genomic value to be used twice as often as the worst individual.
Selection ratio type (leave empty to not change) ⓘ Use this if you want to use some individuals more frequently for reproduction. A value of 2 will lead to the individual will the highest genomic value to be used twice as often as the worst individual.
Index for Selection ratio / OGC ⓘ Select the index used for the selection ratio or OGC.
Number of Repeat (leave empty to not change) ⓘ Number of times the Repeat is executed. The parent node will be the new input of the child node and everything in between is rerun. Nodes will be named the same + _1, _2, _3 etc.
Selection Type ⓘ Random: Each individual is taken with same probability, BVE/Pheno selected based on BVE/Phenotypes
BVE Method ⓘ Direct-Mixed-Model assumes known heritability and thereby is fastest. EMMREML/sommer use the respective R-package. Sommer also support multi-variable models but takes much longer! Bayesian models used are using BGLR. MAS is using lm() on randomly sampled effect markers.
Solving Technique ⓘ Regular Inversion will be done using a Cholesky decomposition and will scale cubically in the number of individuals. For larger datasets other solvers will be more efficient.
Number of markers used for MAS (leave empty to not change) ⓘ This is a very simple implementation of MAS using a linear model of randomly sampled effect markers
Accuracy for Trait {{ trait[["Trait Name"]] }} (leave empty to not change) ⓘ If nothing is entered here the accuracy will be chosen according to the starting heritability
Selection Index ⓘ Select which index to use for BVE/Phenotypic selection
Threshold for selection (leave empty to not change) ⓘ Calculate reliability of BVE according to vanRaden 2008
Threshold sign ⓘ Only keep individuals with higher/lower/eval estimated breeding value
Calculate reliability ⓘ Calculate reliability of BVE according to vanRaden 2008
Estimate reliability ⓘ Estimate the reliability according the the correlation between BVE and BV
Skip traits with no index weight ⓘ Activating this will lead to not performing BVE that are not assign with any weight (=0) in the selection index.
Input Phenotype ⓘ Use the avg. phenotype of the offspring as the phenotype of the individuals in the parent node. Make sure there are offspring! (use of time needed!)
Relationship Matrix ⓘ According to vanRaden 2008, pedigree-based or singleStep (H based on Legarra 2014)
Depth of Pedigree (# generation back) (leave empty to not change) ⓘ Depth of the pedigree. All individuals before are assumed to be unrelated! If you need something else contact Torsten or use the R-package.
Cohorts used in BVE ⓘ Manual select provides you with maximum flexibility to select which individuals to use.
Select cohorts for manuel select: ⓘ Cohorts available for selection are only refresh on reloading or when EDITING an edge. We are working on that.
Sex

Number of Individuals
(leave empty to not change)

Phenotyping Class

Housing Cost Class



Breeding Scheme

For plant breeding the terms 'male' and 'female' should not be seen as strict! Instead interprete them as the 'first parent' and 'second parent' used for reproduction. This is particular needed when multiple cohorts are used for the generation of downstream plants.

Draw your breeding scheme in the following interactive environment.
• Nodes represent cohorts of individuals with similar/same characteristics (e.g. age / sex / genetic origin)
• Edges represent breeding actions that can be taken in a breeding program

• For each breeding action and node you will be provide with a lot more options are generation
• On default only basic options are displayed. In case you are missing something you can most likely activate it
the General information > Advanced Settings > Advanced Edge/Node options
• If not and you think its useful - Contact us!

• Add a Node : Click Edit -> Add Node -> Click in an empty space to place a new node.
• Edit a Node : Select a Node from the diagram -> Click Edit Node
• Drag a Node : Add/Edit a Node or Edge -> Use grey area to drag the Node/Edge display box.
• Double-Click a Node/Edge : Display-only box appears for a node or edge. Need to click Edit Node to edit the node/edge.
• Copy Node : Select a Node -> Right Click.

Legends


Nodes:
▭ {{key}}

Edges:

↗ {{key}}

{{node_operation}}

Drag Node in this Area!

Node {{ active_node.id }}


Copy Node-Options
Name ⓘ Name of the cohort. Please avoid names with : or _ . In particular avoid trailing numbers like ABC_1. Repeated nodes will use this syntax!
Number of individuals ⓘ Number of individuals in this cohort
Ignore Size scaling ⓘ Are the individuals in this cohort founders or are they generated?
Founder ⓘ Are the individuals in this cohort founders or are they generated?
Genotype generation type ⓘ How to generate genotypes/haplotypes for founder individuals
Upload genotypes (plink/vcf) ⓘ You can provide a dataset in plink/vcf format. Only used markers that are included in the map. Make sure your dataset is phased or activate phasing via checking Phasing required!
Path genotypes {{active_node.Path}}
Phasing required ⓘ If imputation via BEAGLE 5.0 default will be automatically processed. This is using the MoBPS function pedmap.to.phasebeaglevcf(). It is not supported when running MoBPS on your own system as paths will usally not match! Contact Torsten if interested to use this locally.
Population allele frequencies ⓘ Pick the originating subpopulation. Allele frequencies for nodes with no genotype data from the same subpopulation will have similar allele frequencies.
Sex ⓘ Sex of the individuals in this cohort
Phenotyping Class ⓘ Choose one of the phenotyping classes generated prior. Default is observation of all traits.
Housing Cost Class ⓘ Choose the housing class for the individuals in this cohort
Array used ⓘ Select the array to use for genotyping. Mostly relevant when a genomic BVE is applied as only markers genotyped in all genotyped individuals will be used.
Proportion of genotyped individuals ⓘ Select the share of genotyped individuals. Mostly relevant for the use of single-step and economic calculations.
Delete cohort information ⓘ Setting this will automatically delete recombination history etc. of the cohort to reduce file size.
Proportion of Male

   

{{edge_operation}}

Drag Edge in this Area!

Edge {{ active_edge.id }}


ⓘ Variable names according to Wellmann et al. 2019: ub.bv (upper bound on genomic value), eq.bv (target genomic value), lb.bv (lower bound genomic value), ub.sKin (upper bound for avg. kinship), uniform (equal contributes of male/females), lb.BV.increase (minimum increase in genomic value), ub.sKin.increase (maximum increase in avg. relationship)
Copy Edge-Options
Breeding Type ⓘ Selected the breeding action you want to use to generate new cohorts and/or link them.
ID of 2.Sub-Group ⓘ Some breeding action require multiple cohorts
Time needed ⓘ Provide the time this breeding action takes. Reproduction currently needs at least one time unit. {{geninfo['Time Unit']}}
Use last available ⓘ Activating this will use the last repeat available from that node - not necessary the node from the same repeat
OGC ⓘ Use optimum genetic contribution according to Wellmann et al. 2019 to determine how often each individual is used for reproduction.
OGC Target ⓘ Variable names according to Wellmann et al. 2019: min.sKin (minimize kinship), max.BV (maximize genomic value), min.BV (minimize genomic value)
OGC Relationship matrix ⓘ Relationship matrix used in OGC
OGC constrain 1
OGC constrain 2
OGC constrain 3
Avoid full-sib matings ⓘ Activate this to not use full-sibling pairs for reproduction.
Avoid half-sib matings ⓘ Activate this to not use half-sibling pairs for reproduction.
Maximal # offspring ⓘ Maximum number of offspring each individual from the parent node is allowed to have.
Maximal # offspring from one mating pair ⓘ Maximum number of offspring from one mating pair.
Selection ratio ⓘ Use this if you want to use some individuals more frequently for reproduction. A value of 2 will lead to the individual will the highest genomic value to be used twice as often as the worst individual.
Selection ratio type ⓘ Use this if you want to use some individuals more frequently for reproduction. A value of 2 will lead to the individual will the highest genomic value to be used twice as often as the worst individual.
Index for Selection ratio / OGC ⓘ Select the index used for the selection ratio or OGC.
ID of 2.parent
Number of Repeat ⓘ Number of times the Repeat is executed. The parent node will be the new input of the child node and everything in between is rerun. Nodes will be named the same + _1, _2, _3 etc.
Selection Type ⓘ Random: Each individual is taken with same probability, BVE/Pheno selected based on BVE/Phenotypes
BVE Method ⓘ Direct-Mixed-Model assumes known heritability and thereby is fastest. EMMREML/sommer use the respective R-package. Sommer also support multi-variable models but takes much longer! Bayesian models used are using BGLR. MAS is using lm() on randomly sampled effect markers.
Solving Technique ⓘ Regular Inversion will be done using a Cholesky decomposition and will scale cubically in the number of individuals. For larger datasets other solvers will be more efficient.
Number of markers used for MAS ⓘ This is a very simple implementation of MAS using a linear model of randomly sampled effect markers
Accuracy for Trait {{ trait[["Trait Name"]] }} ⓘ If nothing is entered here the accuracy will be chosen according to the starting heritability
Selection Index ⓘ Select which index to use for BVE/Phenotypic selection
Selection Proportion ⓘ Selection intensity - the tool will automatically calculate this for you based on the size of the parent/child node. {{ selection_proportion }} (calculated based on # Individuals)
Threshold for selection ⓘ Calculate reliability of BVE according to vanRaden 2008
Threshold sign ⓘ Only keep individuals with higher/lower/eval estimated breeding value
Calculate reliability ⓘ Calculate reliability of BVE according to vanRaden 2008
Estimate reliability ⓘ Estimate the reliability according the the correlation between BVE and BV
Skip traits with no index weight ⓘ Activating this will lead to not performing BVE that are not assign with any weight (=0) in the selection index.
Input Phenotype ⓘ Use the avg. phenotype of the offspring as the phenotype of the individuals in the parent node. Make sure there are offspring! (use of time needed!)
Relationship Matrix ⓘ According to vanRaden 2008, pedigree-based or singleStep (H based on Legarra 2014)
Depth of Pedigree (# generation back) ⓘ Depth of the pedigree. All individuals before are assumed to be unrelated! If you need something else contact Torsten or use the R-package.
Cohorts used in BVE ⓘ Manual select provides you with maximum flexibility to select which individuals to use.
Select cohorts for manuel select: ⓘ Cohorts available for selection are only refresh on reloading or when EDITING an edge. We are working on that.

   


Intension of this module is to aline nodes in the flash application to get nice graphs
• All nodes with similar x or y coordinate will be put to the same value in the respective axis
• Increase values to align more nodes / Reduce values to align less nodes
• Expand flash environment will increase the size of the flash environment

Max X-Axis difference Max Y-Axis difference

Expand flash enviroment1 : ⓘ Please click this button to have bigger breeding scheme diagram and save the project to rearrange nodes and edges.




Analyze Breeding Program with R

Click on button 'Start R Simulation' to run R.

Please check warnings before running the simulation!



Click here to estimate computing time / costs of your breeding program


Or click here to restore the latest R result that you have already generated for the breeding program {{geninfo['Project Name']}}.



{{warningsLog[index]}}

{{warningsLog[index]}}

Download Rdata : ⓘ Download the output of the simulation as Rdata file to save it on your own machine or use it for further analysis in R.    {{geninfo['Project Name']}}.RData


Attention: The actual breeding program must match the uploaded R results, otherwise plotting functions will not work.


Click here to clear all results and plots.





Results: Summary


Download Cohorts List: ⓘ Download CSV file containing cohort names, number of individuals, time-point of generation, total costs, genotyping costs, cost phenotyping and housing costs.  Download {{geninfo['Project Name']}}.csv


Cohort Name Nr. of Individuals Time point Total costs Cost genotyping Cost phenotyping Cost housing
{{cohortsList[index]['Cohort name']}} {{cohortsList[index]['Nr. of individuals']}} {{cohortsList[index]['Time-point']}} {{cohortsList[index]['Total costs']}} {{cohortsList[index]['Cost genotyping']}} {{cohortsList[index]['Cost phenotyping']}} {{cohortsList[index]['Cost housing']}}

Download Cohorts List: ⓘ Download CSV file containing cohort names, number of individuals, time-point of generation, total costs, genotyping costs, cost phenotyping and housing costs.  Download {{geninfo['Project Name']}}.csv


Computing time estimates :

Cohort Name BVE Time Generation Time TOTAL Time
{{cohortsTimeList[index]['Cohort name']}} {{cohortsTimeList[index]['BVEtime']}} {{cohortsTimeList[index]['Gentime']}} {{cohortsTimeList[index]['Totaltime']}}

The following modules allow you to download information on the simulated data
• The population list itself is stored as an RData-object
• In case you are not using miraculix you will not be able to decode underlying genotypes. If that is needed please deactive miraculix in the Advanced settings > miraculix-active

• Before downloading specific data files (VCF / phenotypes etc) press the Prepare data.
• selecting repeat = 0 means the node itself. The first repeat (1) is the second cycle of the breeding program!

Download population list:

Download RData

Download: Population data

Selected datatype:
Selected cohort:
Selected repeat:
Download Ped Download Map Download VCF Download txt











Display 95% Confidence Intervals
Display Legend

Results: Observed Phenotypes





Results: True Breeding Values





Results: Accuracy of Breeding Value Estimation





Principle Component Analysis

Only consider cohorts in repeat nr.: ⓘ Enter a number of a comma separated list of numbers to only consider selected repeats (e.g.: "5", "5,6,7")
PC to consider: ⓘ Enter a number of a comma separated list of numbers to only consider selected repeats (e.g.: "5", "5,6,7")

You need to re-run analysis before behind able to generate the new PCA plot!!!



Results: Relationship and Inbreeding within Cohorts

Only consider cohorts named: ⓘ Test






Results: Major QTLs (Allele Frequency, exp./obs. Heterozygosity)




Cohorts List :

Cohort Name Nr. of Individuals Time point Total costs Cost genotyping Cost phenotyping Cost housing
{{cohortsList[index]['Cohort name']}} {{cohortsList[index]['Nr. of individuals']}} {{cohortsList[index]['Time-point']}} {{cohortsList[index]['Total costs']}} {{cohortsList[index]['Cost genotyping']}} {{cohortsList[index]['Cost phenotyping']}} {{cohortsList[index]['Cost housing']}}

Download Cohorts List: ⓘ Download CSV file containing cohort names, number of individuals, time-point of generation, total costs, genotyping costs, cost phenotyping and housing costs.  Download {{geninfo['Project Name']}}.csv


Computing time estimates:

Cohort Name BVE Time Generation Time TOTAL Time
{{cohortsTimeList[index]['Cohort name']}} {{cohortsTimeList[index]['BVEtime']}} {{cohortsTimeList[index]['Gentime']}} {{cohortsTimeList[index]['Totaltime']}}


JSON-File Viewer:


ⓘ

The current state of the project will be displayed in the Output Area for further inspection and/or editing

ⓘ

Import JSON-project from the Output Area below