Please consider mentioning explicitly in the documentation of scanpy.api.tl.diffmap that "the 0-th column is the steady-state solution, which is non-informative in diffusion maps", as mentioned here: https://github.com/theislab/single-cell-tutorial/blob/master/latest_notebook/Case-study_Mouse-intest

I tried to run Hail with Spark 2.4.4 built for Scala 2.12, and it did not work. It does work with Spark 2.4.4 built for Scala 2.11.

Here's the error I got with Scala 2.12:

Py4JJavaError: An error occurred while calling z:is.hail.HailContext.apply.
: java.lang.NoSuchMethodError: scala/Predef$.refArrayOps([Ljava/lang/Object;)Lscala/collection/mutable/ArrayOps; (loaded from file:/home/hammer

CodeCov looks like a good idea.

Seems really easy as we use Travis already. Are there other solutions?

We used to use emma, but I think that's been lost when we moved to gradle from ant.

Look at https://molgenis.gitbooks.io/molgenis/content/guide-python.html

This is old documentation for the Python Client, back when it was shipped with MOLGENIS. This page should be removed or updated so that it uses the client from molgenis-py-client.

Double check additional installs on Linux with Agg as the Matplotlib backend (9971abe).

Check if libatlas-base-dev is also a required install.

Add Sphinx docs with instructions for installing Python, setting up a virtualenv, etc.

Not sure if this useful for others / including in docs, but here's how I create a conda environment for cmapPy in python 3. I mention this b/c I have done it other ways and ended up not having the hdf5 command line tools in the environment (which I find useful).

0. conda create -n cmapPy3 python=3 scikit-learn scipy numpy seaborn matplotlib statsmodels pandas jupyter sympy h5py
1. `conda ac

There used to be a tutorial button in the menu but it's disappeared, seemingly in the last month or so.

Hi, @hardingnj.
In the old version(2009) of XP-CLR, it distinguish the genotype1 and genotype2 to object population and reference population, respectively. After apply my data in the new version using VCF format, I notice there are only parameters of -samplesA and -samplesB, I'm wondering which samples represents the reference population (A or B)? Since I didn't find the informat

Perhaps it is me who misunderstood, but in the tutorial on survival analysis, we can see this line:

NA06994 569.4273004275149 0 48 20.2946857226 1

Where 569.4273004275149 is the time to event and 0 means that the event was not observed.

If I understand correctly, there shouldn't be a TTE if the event was unobserved and this line would be incorrect.

Hence my proposed f

In order to help others understand how our package works, we have set goals for the way our package is organized, branded, documented, and produces output. This epic will hopefully bring some of those ideas together.

Hi,
umx is great, thank you for it!

I am looking for mediation-relate resources. Are there any? Is it possible to provide an example of mediation analysis with umx? I couldn't find it in the docs, only in the lavaan-to-openmx-helper function.

Particularly, it would be fantastic to exemplify hierarchical (mixed-effects) mediation because I think openmx/umx is the only framework in R that ca