Anyone who is a true advocate of information visualization will be fascinated by this project. Using health records from the Columbia University Medical Center, a group of researchers lead by Andrey Rzhetsky at Columbia University have mapped the overlap between 161 different diseases by studying epidemiological data from 1.5 million patients. Some of the disease correlations found in the mapping exercise have been previously observed, but many are new. The strongest results concern neurological disorders. "We're the first to quantify the overlap between schizophrenia, bipolar, and autism," says Rzhetsky. A Columbia University computer model was used to generate these maps, which the researchers hope will foster further investigation on the genetic bases of the diseases they studied.The first map (top image) is the resulting output of an independent study into one of the 161 diseases mapped in the Columbia study - Migraine. The Columbia researchers found genetic overlap between migraine and about 60 other diseases. The size of the circles corresponds to the relative size of the patient population sampled; in this map, populations range from 46 to 136,000 people. Very little is known about what predisposes a person to migraines, says Rzhetsky, but the mapping project reveals a strong correlation with autism. Rzhetsky's research also shows correlations between infections and many neurological diseases, including migraines, autism, and schizophrenia.
The second map (bottom image) shows the overlap between some common diseases. Red lines between diseases indicate a positive correlation; blue lines indicate a negative correlation. The thickness of a line corresponds to the strength of the correlation. The size of the circles corresponds to the relative size of the patient population for each disease; in this map, patient populations range from about 20 to about 136,000 people. This visualization also demonstrates a strong positive correlation between schizophrenia, bipolar disorder, and autism (yellow circles), suggesting that they are caused by a common group of genes.
