It's the most common cause of childhood physical disability, affecting as many as three children out of every 1,000 babies born around the world each year.
But despite the prevalence of cerebral palsy, scientists don't fully agree on how it emerges, or even what the condition really is.
Historically, the movement disorder was attributed to complications at birth, including prematurity, infection, and lack of oxygen.
In recent years, though, alternative explanations have surfaced, with a number of studies suggesting cerebral palsy may be a genetic disorder linked to pathogenic variants of individual genes.
But a new study published in the American Journal of Human Genetics suggests the truth is quite different.
"We wanted to determine how strong the evidence really is linking specific genes to cerebral palsy," researchers Peter Robinson and Adam Arterbery from the Jackson Laboratory for Genomic Medicine in Farmington, Connecticut told ScienceAlert via email.
"We expected that some genes reported as causal for cerebral palsy would have stronger evidence than others, and that a more systematic analysis would distinguish well-supported associations from potentially incidental findings."
For their study, Robinson, a computational biologist, and Arterbery, a biocuration researcher in Robinson's lab and one of the first authors of the new paper, conducted a sweeping literature review.
With their team, they reviewed 21 published genetic studies of cerebral palsy, identifying 515 genes in total that had previously been reported as potentially causal for the condition.
They then cast a wider net, looking for evidence of disease-causing gene mutations in more than 5,500 other studies, and used statistical analysis to identify which of the 515 suspected genes previously linked to cerebral palsy could plausibly be associated with the condition.
The results were stark.
"The magnitude of the discrepancy became striking when we found that only 89 of the 515 genes previously reported as causal in cerebral palsy had sufficient statistical evidence for cerebral palsy association," Robinson and Arterbery told ScienceAlert.
In a separate phase of their study, the researchers performed whole-genome sequencing on 460 children with cerebral palsy, identifying pathogenic or likely pathogenic variants in 60 genes in about 16 percent of the children.
But only 16 of those 60 genes showed significant evidence of an association with cerebral palsy, per the researchers' literature analysis.
In other words, while genetic testing can identify large numbers of gene variants in children with cerebral palsy, the researchers suggest that some of these previously proposed genetic links may not reflect a true disease-causing mechanism.
Instead, it's possible, the team thinks, that cerebral palsy might not be a single disease or genetic disorder, but rather that it represents a collection of symptoms that occur or overlap with other kinds of conditions.
"We propose here that cerebral palsy can best be understood as a feature of certain Mendelian diseases rather than a separate disease," the researchers write in their paper.
"Our approach is based on a paradigm that considers cerebral palsy to be a disease component (phenotypic feature) rather than a precise disease diagnosis."
The researchers acknowledge that their study does not mean that hundreds of these gene variants don't cause cerebral palsy, or don't have links to the condition.
Related: Everyone Assumes DNA Shapes Your Personality. It Turns Out Something Else Is Doing Most of The Work
Instead, the analysis just suggests that, right now, the amount of evidence we have for many of these candidates is lacking, so examining how cerebral palsy might overlap with other diseases could be a better way of identifying where its various symptoms come from.
"We are excited about the possibility of moving beyond simply identifying genetic variants toward understanding how genotype, phenotype and disease mechanisms fit together," the researchers said.
"This could improve the interpretation of genomic testing and ultimately help identify patients who may benefit from more precise evaluation and care."
The findings are reported in the American Journal of Human Genetics.