I have thousands of Fst-values (markers) spread across the genome. I would like to use Loess to visualize and integrate them along the chromosomes. This makes sense only along the chromosome, since markers (and thus Fst values) are physically linked when located in close physical proximity. The problem arises, because the length of different chromosomes varies (the longest is 3-4 times as long as the shortest). If I understand correctly, when I define in loess() "span" (alpha), I define the number of Fst-values AS FRACTION OF ALL Fst-values that belong to the same chromsome when when calculating the local estimated value with loess. E.g., when I take a higher value for "span", a larger percentage of all values will be considered. The problem is, that larger chromosomes have more markers and thus more Fst-values. I therefore expect a bias between large and short chromosomes. Is this true? - If so, is there an option to define "span" (alpha) dependent on the number of values?
thanks -- View this message in context: http://r.789695.n4.nabble.com/Loess-smoothing-alpha-Parameter-tp4307000p4307000.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.