🔬
GWASTic Documentation
  • 👋Welcome to GWAStic documentation
  • Overview
    • 💡Video Tutorials
    • ✨Features
  • Fundamentals
    • 🖥️Installing GWAStic
    • 🖥️Starting GWAStic
    • ↪️Converting VCF to BED files
    • 🔄GWAS Analysis
    • 🔄Genomic Prediction
    • Running from command line
    • Algorithms
    • Effect size and P-value
    • Train, Test, and Validation Sets
    • Settings
    • ℹ️References
    • Version history
Powered by GitBook
On this page
  1. Fundamentals

Settings

PreviousTrain, Test, and Validation SetsNextReferences

Last updated 10 months ago

General settings:

  • Number of jobs: Number of jobs specifies the number of CPU cores to use, with -1 using all available cores, 1 using a single core, and 8 using eight cores for parallel processing.

  • Gigabytes of memory the run: Gigabytes of memory the run should use. If 0, will read the SNPs in blocks the same size as the kernel, which is memory efficient with little overhead on computation time.

  • Maximum number of SNPs: Sets the maximum number of SNPs to be used for the Manhattan and QQ plots. Recommended for large SNP datasets to improve plotting performance, as handling a high number of SNPs can significantly slow down the plotting process. Leave the field empty to use all SNPs (default).

  • Advanced plotting: Enable this setting to access more advanced plotting features for both phenotypic and genotypic statistics. This is particularly useful for creating high-quality graphics suitable for publication. Please note that enabling these options may significantly increase processing time, especially with large datasets. By default, this feature is disabled to optimize performance.

Machine Learning settings:

  • P-value threshold: Enter the significance level for statistical tests. A lower p-value indicates stronger evidence against the null hypothesis.

  • Training size: Set the percentage of the dataset to be used for training the model. The rest will be used for testing.

  • Aggregation method: Choose an aggregation method to combine SNP effects from multiple models (GWAS analysis only).

    • Sum: Adds up the effect from all models.

    • Median: Use the middle value of the effects.

    • Mean: Calculates the average of the effects.

  • Number of trees: Specify the number of trees to be used in the forest. More trees can increase accuracy but also computation time.

  • Max depth: Determine the maximum depth of the trees. Deeper trees can model more complex relationships.