Zika virus shows up in urine, saliva and semen. And it seems to go right to the brain in fetuses, but nowhere else in their bodies.
So it's no wonder this once-obscure virus is mystifying even the experts.
Zika is spreading explosively across Latin America and the Caribbean. The islands of Bonaire and Aruba are the latest to report local spread, according to Marcos Espinal of the Pan American Health Organization.
Such rapid movement is not unexpected when a new virus — in this case, spread by an extremely efficient mosquito — shows up in a population with almost no immunity.
But almost everything else about Zika is surprising — especially the seemingly new links with birth defects and a rare condition called Guillain-Barre syndrome.
"I am trying to understand why what we are seeing in Brazil we haven't seen in 30 years of circulation in Southeast Asia (and elsewhere)," said Dr. Stephen Thomas, an Army colonel and virus expert at the Walter Reed Army Institute for Research.
Doctors largely ignored Zika because it usually caused such mild symptoms, if it even caused any symptoms at all.
One study suggests 80 percent of people infected never even noticed, while the other 20 percent mostly suffered rash and a mild fever.
But in Brazil, it's not only making people noticeably sick, it's suspected of causing a giant increase in serious birth defects and perhaps Guillain-Barre syndrome.
One thing that's holding up research is the lack of a quick and easy test for Zika. Quick tests get Zika mixed up with related viruses that are also common and spreading in the region, including dengue virus and chikungunya.
To try to figure out if anyone else had noticed this, Thomas took the tried-and-true tack of checking the literature — searching through the few scientific papers that have been written about Zika since its discovery in Uganda in 1947.
He found a lot of virus experts have also been baffled by Zika's weird qualities.
For instance, Didier Musso of the Louis Malarde Institute in Tahiti and his colleagues found that Zika was easier to detect in saliva than in blood during an outbreak there in 2013 and 2014. They found Zika in the blood of 28 percent of patients with symptoms, but they could get genetic material from the virus out of 57 percent of saliva samples.
"It makes me question whether blood is the proper fluid to be sampling if (you) want to capture Zika," Thomas told a packed seminar on Zika at the Johns Hopkins Bloomberg School of Public Health this week.
Brazilian researchers have also reported finding evidence of Zika virus in the urine of patients.
Ann-Claire Gourinat of the Institut Pasteur in Noumea, New Caledonia, reported this last year, as well, after studying the Tahiti outbreak.
"Urine samples were positive for Zika virus more than 10 days after onset of disease, which was a notably longer period than for serum samples. This finding supports the conclusion that urine samples are useful for diagnosis of Zika virus infections," they wrote in a report in the journal Emerging Infectious Diseases.
Zika has also been found in amniotic fluid, which mostly is made from the urine produced by the fetus, said Dr. Jeanne Sheffield, director of maternal-fetal medicine at Johns Hopkins.
There's no evidence that the genetic material from Zika that can be found in saliva and urine can infect people. But a handful of reports suggest the mosquito-borne virus can also be found in and transmitted by semen.
But the findings suggest that testing blood for Zika might not help doctors get on top of the epidemic, Thomas said.
"Between the saliva and the semen and the urine, I'm not sure what the best samples might be," he said, adding, "It also makes me question whether this (is) just a mosquito-to-human transmission."
Experts took up the issue at a workshop sponsored by the National Academy of Medicine this week.
"There's an urgent need to understand the other modes of transmission and to communicate effectively about them," Dr. Andrew Pavia, an expert on infectious diseases at the University of Utah, said at that meeting.
Officials at the U.S. Centers for Disease Control and Prevention have not directly answered questions about the case of a person in Dallas who caught Zika from a sexual partner who'd traveled to Venezuela.
They said they presume it was sexually transmitted but reiterated they were investigating further.
One comforting finding is the lack of evidence that mothers can infect their babies through breast milk — although genetic material from Zika can be found in the liquid. The human immunodeficiency virus that causes AIDS can and does infect babies through breast milk and it's very commonly transmitted in semen, but not via saliva or urine.
Another odd effect of Zika — or suspected effect — is what it does to a developing fetus's brain.
It's the No. 1 suspect behind a rise in cases of microcephaly, an often devastating birth defect marked by an underdeveloped brain and smaller-than-normal head.
Doctors who have been able to autopsy miscarriages, abortions or babies who died soon after birth are finding unusual and troubling brain damage not usually associated with microcephaly.
"It's not just a small head. There are a lot of abnormalities in the brain," Sheffield told the seminar.
These include a smoother-than-normal brain surface — a healthy human brain is very wrinkled — and shrunken brain stems. There's evidence of damaged eyes, too.
Other viruses are known to damage the brains of developing infants. A herpes virus called cytomegalovirus and rubella, or German measles, are the most notorious.
But Dr. Albert Ko of Yale University, who's been working in Brazil for decades, says the recent microcephaly cases look different.
"These are severe presentations," Ko told the seminar. Brain scans show calcifications — hardened tissue that has been damaged and died. There are entire missing areas, sometimes replaced by fluid.
And the other viruses usually affect other areas of the baby's body. The Zika-linked damage, however, is confined entirely to the brain.
Thomas says this fits in with decades-old research on animals that suggests Zika prefers to infect nerve tissue.
And that fits in with the final baffling effect of Zika. Is it causing Guillain-Barre syndrome, an inflammation of the nerves that can cause sometime-serious paralysis after any number of infections?
It could be the virus has changed in some way. It could be cross-reacting with its relative, dengue, in people who have been infected with both. Or it could be something else.