type: gene_score
filename: pNull.csv
scores:
  - id: pNull
    desc: Probability of biallelic loss-of-function intolerance
    histogram:
      type: number
      number_of_bins: 126
      view_range:
        min: 0
        max: 1
      x_min_log: 0.00001
      x_log_scale: true
      y_log_scale: true
  - id: pNull_rank
    desc: Gene rank after sorting by pNull score
    histogram:
      type: number
      number_of_bins: 150
      view_range:
        min: 1
        max: 18225
      x_log_scale: false
      y_log_scale: false


meta:
  summary: |
    Probability of LoF tolerance

  description: |
    
    The probability of being null (pNull) reflects a gene’s tolerance to loss-of-function (LoF) variants in both heterozygous and homozygous states. Genes with high pNull values show LoF variants at expected frequencies, indicating little or no selective constraint. Such genes are considered LoF-tolerant, meaning that complete loss of function is typically compatible with normal human health. The pNull score is computed alongside pLI and pRec using population-level LoF data from ExAC and gnomAD.    
    
    [Lek et al., Analysis of protein-coding genetic variation in 60,706 humans, Nature 2016](https://www.nature.com/articles/nature19057)

    Downloaded on: 10/24/25

    Dataprep.py takes the pNull column from Supplementary Table 13 "Gene Constraint" sheet and adds a pNull_rank column.