Table 1.
Limitations of genetic association studies in complex multifactorial diseases such as SRLV-induced pathologies. Factors leading to confounding that may be encountered as well as their consequences are shown (information compiled from [41,42,43,44,45,46,47,48,49].
| HOST | GENOTYPING | ||||
|---|---|---|---|---|---|
| Population stratification | Often present in livestock due to breeding practices | Bias/Partiality | Case samples are treated with preference | ||
| Sample size | Affects power to detect association | Marker density | Often few markers are analyzed per gene | ||
| Phenotype | Description must be accurate | Marker frequency | Frequency of marker variants affect odds of detecting association | ||
| Age | Older animals have been exposed for a longer time | STATISTICAL ANALYSIS | |||
| Gene effect | Genes involved have small/moderate effects | Power | Depends on sample size, marker frequency and gene effect | ||
| Presence of other diseases | May facilitate SRLV pathogenesis | Gene interaction | Often unaccounted for | ||
| SRLVs | Confounding factors | Failure to account for them may lead to erroneous interpretations | |||
| Different virus strains | Different virulence and host/organ spectrum complicates research; | Multiple corrections | Necessary yet may lead to reject real associations | ||
| Strain variability may affect detection of infected individuals | Consistency/replicability | Results must be replicated in different populations/ | |||
| ENVIRONMENTAL FACTORS | Can results be replicated in a different population? | ||||
| Husbandry | Prolonged and crowded housing enhances infection | ||||