@@ -491,7 +491,7 @@ the definition for the mitochondrion is fully contained within the melanosome me
In this [example](https://gist.githubusercontent.com/richardtjornhammar/e84056e0b10f8d550258a1e8944ee375/raw/e44e7226b6cb8ca486ff539ccfa775be981a549c/example9.py) we do exactly this for two gaussian point clouds. The dbscan search is just a single line `dbscan ( data_frame = point_cloud_df , eps=0.45 , minPts=4 )`, while the last lines are there to plot the [results](https://bl.ocks.org/richardtjornhammar/raw/0cc0ff037e88c76a9d65387155674fd1/?raw=true)(has[graph revision dates](https://gist.github.com/richardtjornhammar/0cc0ff037e88c76a9d65387155674fd1/revisions) )
The [radial distribution function](https://en.wikipedia.org/wiki/Radial_distribution_function) is a useful tool for visualizing whether or not there are radial clustering tendencies at any average distance between the group of interest and any other constituents of the system. This structure assessment method is usually used for [analysis](https://gist.githubusercontent.com/richardtjornhammar/33162d3be1e92f1b1fafbd9e46954e91/raw/18b004d3dd8a8a83af0f15c3f4aa3927dc62e5f8/argon.py) of particle systems, i.e. see [liquid structure](https://bl.ocks.org/richardtjornhammar/raw/bc1e9a8b4c693a338ef812a74ab685e9/?raw=true). It is implemented in the `clustering` module and is demonstrated [here](https://gist.githubusercontent.com/richardtjornhammar/f25ec2eef0703f07ebc0d678123f450e/raw/96a8e0d3fb6be03e0ea3e35b5470ae35513f5142/example10.py). If there is a significant density close to `r=0` then you cannot separate the group from the mean background. This also means that any significance test between those groups will tell you that the grouping is insignificant. The [resulting plot](https://bl.ocks.org/richardtjornhammar/raw/ff417450790c8c885b077fc7ee20409d/?raw=true) has [revision dates](https://gist.github.com/richardtjornhammar/ff417450790c8c885b077fc7ee20409d/revisions). Since the radial distribution function calculates the spherically symmetric distribution of points surrounding an analyte, or analyte group, of interest it is effectively analogous to segmenting the distance matrix and leaving out any self interaction distances that may or may not be present.
The [radial distribution function](https://en.wikipedia.org/wiki/Radial_distribution_function) is a useful tool for visualizing whether or not there are radial clustering tendencies at any average distance between the group of interest and any other constituents of the system. This structure assessment method is usually used for [analysis](https://gist.githubusercontent.com/richardtjornhammar/33162d3be1e92f1b1fafbd9e46954e91/raw/c0685bb79527c947213ffe08973d3ea4e072257e/argon.py) of particle systems, i.e. see [liquid structure](https://bl.ocks.org/richardtjornhammar/raw/bc1e9a8b4c693a338ef812a74ab685e9/?raw=true). It is implemented in the `clustering` module and is demonstrated [here](https://gist.githubusercontent.com/richardtjornhammar/f25ec2eef0703f07ebc0d678123f450e/raw/b9ac597a9d2587727af3cb06a8090ad0eaf0ba49/example10.py). If there is a significant density close to `r=0` then you cannot separate the group from the mean background. This also means that any significance test between those groups will tell you that the grouping is insignificant. The [resulting plot](https://bl.ocks.org/richardtjornhammar/raw/ff417450790c8c885b077fc7ee20409d/?raw=true) has [revision dates](https://gist.github.com/richardtjornhammar/ff417450790c8c885b077fc7ee20409d/revisions). Since the radial distribution function calculates the spherically symmetric distribution of points surrounding an analyte, or analyte group, of interest it is effectively analogous to segmenting the distance matrix and leaving out any self interaction distances that may or may not be present.
The functions `select_from_distance_matrix` uses boolean indexing to select rows and columns (it is symmetric) in the distance matrix and the `exclusive_pdist` function calculates all pairs between the points in the two separate groups.
