EIGENSTRAT vs MDSOutlier: Genetic Background Outlier Detection

Question

[Shicheng Guo]

List for Genetic Background Outlier Remove Programme

Anyone compared these two tools before? EIGENSTRAT vs mdsoutlier

Which one is better? What's the difference between the results?

First: EIGENSTRAT

https://github.com/Shicheng-Guo/EIG/tree/master/EIGENSTRAT

Second: mdsoutlier

https://sourceforge.net/projects/mdsoutlier/

[Kevin Blighe]

Your links were broken, which is probably why nobody responded for more than a week. I have just fixed them.

My general rule of thumb is >3 standard deviations and it's an outlier. Outlier detection is not and should never be an exact science, though. There are many situations in which an outlier can be due to genuine biological reasons, meaning that by eliminating it you'd be throwing away an interesting sample.

If EIGENSTRAT and mdsoutlier both have publications behind them, then you can easily justify the use of either.