BanyanDNA and the Habsburgs

Who can pass up a chance to mix genealogy with science, especially with a European vacation thrown in?

Last spring, I visited Schönbrunn Palace in Austria, the summer residence of the Habsburg dynasty.  That got me thinking.

 

A Natural Experiment

The Habsburgs’ renown as European kings and emperors is matched by their infamously consanguineous marriages.¹  Uncles married nieces.  First cousins wed, as did double first cousins.  Spouses could be related in dozens of ways.

These marriages were contracted to consolidate and maintain power within the family, but there was a down side:  centuries of inbreeding resulted in high infant and child mortality rates as well as physical and mental disabilities in adults.  The so-called “Habsburg jaw” was an inherited deformity of the lower jaw that made biting, chewing, and speaking difficult.  King Charles II of Spain reportedly had persistent digestive problems as a result.  For that and his many other ailments, he was nicknamed “El Hechizado,” or the Bewitched.  He was apparently sterile and, despite two marriages, died childless at age 38, sparking the 13-year War of Spanish Succession.

Scientists have a term for the harmful effects of consanguinity:  inbreeding depression.  It refers not to a psychological state but to the decline in survival and fertility that results when offspring inherit the same harmful gene variants from both parents.

We all carry harmful gene variants.  Fortunately, we inherit two complete sets of autosomal DNA—one from each parent—so we have a backup.  That is, if you inherit a bad version of a gene from dad but a good one from mom, you’re usually fine.  With inbreeding, though, the chances of inheriting two bad copies increases drastically.

The long-term effects of Inbreeding depression are challenging to study in humans.  Ideally, we’d design a controlled experiment with two pools of research subjects, one randomly assigned to inbreed and the other to outbreed.  There are two problems with that.  First, a robust comparison would take many generations and second, it would be highly unethical.

The Habsburgs offer instead what is called a natural experiment:  conditions that happened on their own (rather than by experimental design) yet provide scientifically useful data.  Unlike other historical dynasties, like the Egyptian pharaohs, that practiced consanguineous marriages, the genealogy of Habsburgs is well documented, as are their fertility and mortality rates.  They make for an ideal study.

Scientists at the University of Santiago de Compostela in Spain did just that:  they analyzed the effects of inbreeding on the Habsburg Holy Roman Emperors and Spanish kings.²  Using a tree with more than 4,000 individuals spanning more than 20 generations, they found that the degree of inbreeding had a strong negative effect on both infant survival and childhood survival.

As for Charles II, although his parents were uncle and niece, they shared as much DNA as full siblings because they were related in so many other ways as well.

(23andMe published an interesting blog on this Habsburg study.)

 

Hmmmmm!

I wondered if I could replicate part of this study in BanyanDNA, a tool designed to analyze DNA matches in complex genealogies.  BanyanDNA simulates the inheritance of DNA through a tree you provide, so it is completely customizable to a family even as intertwined as the Habsburgs.

First, I built a partial tree of the Habsburgs in BanyanDNA (equivalent to Figure 1 in the Ceballos paper).  This tree had 56 individuals and 15 consanguineous marriages.

Then, I tricked BanyanDNA into thinking those couples had done DNA testing by telling it that each husband shared 1 cM of DNA with his wife.  (One centimorgan is a silly amount of DNA, of course.  The point is to make BanyanDNA treat the couples as DNA testers.)

Finally, I ran a validation analysis in BanyanDNA.  The tool uses a sophisticated computer model to simulate the inheritance of DNA from the ancestors down through the tree.  Then for each pair of “testers,” it estimates how much DNA would be expected to share given their relationship(s).

It even tallies all of the relationships for you.  Philip IV of Spain and his wife Mariana of Habsburg were related at least 64 unique ways.

Here’s what BanyanDNA found.  The couples are listed from left to right and from top to bottom in the tree.

The couples Philip IV of Spain & Mariana of Habsburg and Leopold I of Habsburg & Margaret Theresa of Spain (in red) were each uncle–niece but shared as much DNA as would full siblings because they were related in so many other ways.  Charles II the Bewitched was the child of Philip IV & Mariana.  His sister Margaret Theresa married their uncle, to whom she was related at least 189 different times!

My analysis, of course, is an underestimate of the relationships, because I only used 56 people whereas the full family tree contained more than 4,000.  Even so, it’s exciting to build the tree myself and replicate the work of the Spanish scientists.

Best of all, you can do this, too!  If you have ancestors who were related multiple times, and you’re curious how much DNA they might have shared with one another, you can build their tree in BanyanDNA and do a similar analysis.

Join the BanyanDNA User Group on Facebook to share what you learn or to get pointers on using the tool.

---

¹ The word “consanguineous” means “of the same blood” or “biologically related.”  It comes from the Latin roots con, meaning together, and sanguineous, referring to blood.  A consanguineous marriage is a marriage between relatives, usually 2nd cousins or closer.
² Ceballos FC & G Álvarez (2013) Royal dynasties as human inbreeding laboratories: the Habsburgs. Heredity 111: 114–121. doi:10.1038/hdy.2013.25