Opportunistic untargeted sequencing can complement targeted viral surveillance workflows for unresolved respiratory illness in Uganda

Every year, millions of people around the world catch colds and flu-like illnesses that send them to doctors or keep them at home. In many places, especially where medical resources are stretched thin, health workers rely on specific tests to find out what is making a person sick. These tests are like looking for a few specific keys in a giant keyring; they are excellent at finding the most common culprits, such as the flu virus or the virus that causes COVID-19. However, because these tests only look for what they are designed to find, they often miss the other viruses that might be hiding in the same sample. When a patient has a fever, a cough, and a sore throat, but the standard tests come back negative, the cause of their illness often remains a mystery. This leaves a gap in our understanding of respiratory health and can lead to unnecessary treatments, such as antibiotics that do not work against viruses.

To fill this gap, scientists are exploring a different approach called untargeted sequencing. Instead of hunting for one specific virus, this method reads the genetic code of everything present in a sample, much like reading every book in a library to see what stories are there, rather than just checking the shelves for a single title. This technique allows researchers to spot viruses they were not even looking for. A new study from Uganda demonstrates how this broad approach can work alongside standard testing to solve medical puzzles that would otherwise remain unsolved.

In December 2025, a team of researchers at the Uganda Virus Research Institute collected samples from four adults in the Wakiso District who were suffering from flu-like symptoms. All four individuals had been tested with the standard methods used in routine surveillance, and all four had tested negative for both influenza and SARS-CoV-2. Despite the negative results, the patients still felt unwell, with symptoms ranging from fever and headache to chest pain and difficulty breathing. The researchers decided to apply an opportunistic, untargeted sequencing method to these four specific samples. They extracted the genetic material from the nasal swabs and used a powerful sequencing machine to read the genetic code of every virus and bacterium present, without making any assumptions about what they might find.

The results were revealing. While two of the four samples showed no signs of a respiratory virus, the other two contained clear evidence of infection by viruses that the standard tests had missed. In one case, the sequencing uncovered a nearly complete genetic blueprint of a virus called Human Metapneumovirus. This virus is a known cause of respiratory illness worldwide but is not typically included in routine testing panels in many regions. The researchers were able to reconstruct almost the entire genome of the virus, a string of genetic code measuring 13,273 base pairs, with high clarity. By comparing this genetic sequence to others stored in global databases, they found that this virus belonged to a lineage known as B2, which circulates widely around the world. The analysis also showed that while most of the virus's genes were very similar to those found in other parts of the world, one specific part of the virus, responsible for helping it attach to human cells, showed more variation than the rest.

In a second case, the sequencing detected a different virus, known as Human Respirovirus 1, which is also a common cause of respiratory disease. However, the signal for this virus was much weaker. The researchers could only recover a small fragment of its genetic code, a piece measuring 368 base pairs. This fragment was enough to confirm the presence of the virus, but the low amount of genetic material available meant they could not reconstruct the full genome. The difference in the amount of genetic material found between the two cases suggests that the first patient had a higher amount of virus in their system, making it easier to detect, while the second patient likely had a lower viral load or the virus had begun to break down, making it harder to capture a complete picture.

The study also highlighted what happens when the method finds nothing. In the two samples where no respiratory virus was detected, the sequencing still picked up bacteria that naturally live in the human nose and throat, such as Prevotella melaninogenica and Corynebacterium propinquum. The presence of these common bacteria confirmed that the samples were valid and that the sequencing process was working correctly. The absence of a virus in these two cases does not necessarily mean the patients were not sick; it simply means that if a virus was present, it was either at a level too low for this method to catch or the illness was caused by something other than a virus. The researchers noted that this technique is less sensitive than the standard tests designed to find specific viruses, so a negative result here does not rule out an infection entirely.

This work serves as a proof of concept for using broad, untargeted sequencing as a tool to complement routine surveillance. The researchers did not set out to track how common these viruses are in the population, nor did they test a large number of people. Instead, they took a small, targeted look at a group of patients whose illnesses had gone unexplained by standard methods. The success of finding two distinct viruses in just four samples suggests that this approach can be a valuable addition to the diagnostic toolkit. It offers a way to uncover the "missing" causes of respiratory illness, providing a more complete picture of what is circulating in the community. By occasionally applying this broad-spectrum view to cases that standard tests cannot solve, health officials in Uganda and similar settings can gain a deeper understanding of the viral landscape, ensuring that fewer patients are left without a diagnosis.