Paperonium

HIV research · PLoS Pathog · 2013

One sentence says HIV patients have more cAMP, a chemical, then prints both groups’ averages swapped.

One sentenceThe paper’s chart548.3People with HIV760.5People with HIVread off its dots761.6Healthy people547.8Healthy peopleread off its dots
One sentenceThe paper’s chart548.3People with HIV760.5People with HIVread off its dots761.6Healthy people547.8Healthy peopleread off its dots

Average cAMP per million immune cells, in femtomoles (a count of molecules)

The paper says

Immune cells from people with HIV contain more cAMP, a chemical that holds them back, than healthy people’s, the paper says. One sentence calls patients’ level “higher”, then gives patients 548.3 and healthy people 761.6, in femtomoles (a count of molecules) per million cells.

Its own data show

The paper’s own chart shows each person as a dot. Patients’ dots run from about 623 to 819, yet the sentence gives them an average of 548.3. Read off the chart to within about 1 femtomole, the dots average 760.5 for patients and 547.8 for healthy people: the sentence’s two numbers, on the wrong groups.

An average always lies between a group’s lowest and highest values, so the dots settle it: patients have more cAMP, as the paper says. That finding stands; the error is only in this sentence’s numbers. But anyone copying them, say to compare studies, would record patients as 28% lower instead of 39% higher.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

Regulatory T Cells Negatively Affect IL-2 Production of Effector T Cells through CD39/Adenosine Pathway in HIV Infection · PLoS Pathog (2013) · doi:10.1371/journal.ppat.1003319