Metrics
Category Metrics
Overview
Summary fingerprinting metrics and statistics about the comparison of the sequence data from a single read group (lane or index within a lane) vs. a set of known genotypes for the expected sample. @author Tim FennellThis table summarizes the values that are specific to this metric.
| Metric | Summary |
|---|---|
| READ_GROUP |
The read group from which sequence data was drawn for comparison. |
| SAMPLE |
The sample whose known genotypes the sequence data was compared to. |
| LL_EXPECTED_SAMPLE |
The Log Likelihood of the sequence data given the expected sample's genotypes. |
| LL_RANDOM_SAMPLE |
The Log Likelihood of the sequence data given a random sample from the human population. |
| LOD_EXPECTED_SAMPLE |
The LOD for Expected Sample vs. Random Sample. A positive LOD indicates that the sequence data is more likely to come from the expected sample vs. a random sample from the population, by LOD logs. I.e. a value of 6 indicates that the sequence data is 1,000,000 more likely to come from the expected sample than from a random sample. A negative LOD indicates the reverse - that the sequence data is more likely to come from a random sample than from the expected sample. |
| HAPLOTYPES_WITH_GENOTYPES |
The number of haplotypes that had expected genotypes to compare to. |
| HAPLOTYPES_CONFIDENTLY_CHECKED |
The subset of genotyped haplotypes for which there was sufficient sequence data to confidently genotype the haplotype. Note: all haplotypes with sequence coverage contribute to the LOD score, even if they cannot be "confidently checked" individually. |
| HAPLOTYPES_CONFIDENTLY_MATCHING |
The subset of confidently checked haplotypes that match the expected genotypes. |
| HET_AS_HOM |
num of hets, observed as homs with LOD > threshold |
| HOM_AS_HET |
num of homs, observed as hets with LOD > threshold |
| HOM_AS_OTHER_HOM |
num of homs, observed as other homs with LOD > threshold |
GATK version 4.6.0.0-33-gdffedfb built at Wed, 23 Oct 2024 21:44:48 -0400.
0 comments
Please sign in to leave a comment.