Genomic variant annotation and conservation scoring · BEGINNER LESSON
GERP++
GERP++ is an evolutionary conservation-based score used to estimate how constrained a genomic position is across species. Sites with higher scores are taken to be more likely under selective constraint, so variants there are more likely to be deleterious.
What GERP++ is and what it measures
GERP++ is a method that uses evolutionary conservation to score genomic positions. It measures how unusual a substitution is across a phylogeny, which is a branching history of related species. If a site changes rarely across many species, that suggests strong constraint, meaning natural selection has tended to preserve that position; variants at such sites are then predicted to be deleterious more often than variants at weakly constrained sites. In one study, GERP scores were computed from an alignment of 46 species, including 16 reptiles, showing that the method works from a multi-species comparison rather than from a single genome alone.
How the score is used
The sources describe GERP++ as a conservation-based approach for estimating genetic load, which is the burden of harmful variation in a population. In practice, a mutation can be classified as putatively deleterious when its GERP score is above a chosen threshold; one cited analysis used a cutoff higher than 4. Other studies combine GERP++ with other variant features, such as stop-gain status, predicted protein effect, and whether the site lies in a functional domain. This shows that GERP++ is often one input in a larger filtering or prioritization scheme, not the only piece of evidence.
[1] pmc.ncbi.nlm.nih.gov · articles · PMC13219662The relationship between genomic variation and genetic load:... [2] Table 3.
Mechanism and interpretation
The core idea is simple: if a DNA position has stayed similar across many species, evolution has likely been resisting change there. GERP++ turns that pattern into a score, so higher values mean stronger inferred constraint. This makes the score useful for finding variants that may matter biologically, including noncoding variants, which are DNA changes outside protein-coding regions. The sources also note that GERP++ is commonly discussed alongside other conservation metrics such as PhyloP and PhastCons, but it is one member of a broader group of conservation-based tools.
[1] pmc.ncbi.nlm.nih.gov · articles · PMC13219662The relationship between genomic variation and genetic load:... [3] Evolving computational paradigms for noncoding variant ...
Limitations and cautions
GERP++ is not a direct experiment measuring fitness. The sources warn about false positives, especially when thousands of single-nucleotide polymorphisms, or SNPs, are analyzed. A false positive is a variant flagged as harmful even though it is not truly harmful. The cited discussion says that GERP scores have correlated with experimentally measured fitness effects in humans and dogs, but also that more standardized metrics and functional studies, either in living systems or in the lab, are needed to confirm the biological impact of variants. In other words, GERP++ is supportive evidence, not final proof.
[1] pmc.ncbi.nlm.nih.gov · articles · PMC13219662The relationship between genomic variation and genetic load:... [3] Evolving computational paradigms for noncoding variant ...
SEE THE IDEA
How GERP++ works
Compare the same genomic site across many related species.
Rare substitutions suggest strong evolutionary constraint.
Assign a value that summarizes how constrained the site appears.
Higher scores can help flag variants as more likely deleterious.
Each step feeds the next: species comparison reveals conservation, conservation becomes a score, and the score helps prioritize variants.
GERP++ converts multi-species conservation patterns into a site-level score used to infer selective constraint and support variant interpretation.
TRY IT YOURSELF
Test your understanding
Choose an answer, then check the explanation.
PUT IT TOGETHER
Match each concept to its meaning
Drag a label onto its description, or select a label and then select a description. This also works with a keyboard.
4 pairs to match
Sources and further reading
This general lesson was generated from public source material. It explains a concept; it does not interpret a private paper or provide medical advice.