Other JuliaNeuroscience packages from here

[Tokazama]

There are several packages that I think belong in JuliaNeuroscience. Some of these are just ideas and others I've been waiting for NeuroCore.jl to get in place first. All names should be considered suggestions at this point (not final).

• NeuroGraphs: Graph theory/connectome stuff. I have a lot of code already put together for this but just need to finalize the array interface here.
• NeuroSignals: There needs to be a package for processing signals (which are usually time series related). I know there's a strong ERP signal processing community but a lot of the methods are similar to fMRI signal processing. If possible it would be nice to unify this.
• NeuroDatadeps: Interface for neuroscience related datasets a la Datadeps.jl. The difference between this and just using Datadeps.jl would be additional enforcing criteria on how the data is formatted. It would allow users to have some degree of consistency so things like meta analyses would be more straight forward. There are multiple steps to implementing something like this and it would likely take several people to do it right, so if people are interested in this it should probably become a dedicated discussion before it becomes a package.
• NeuroStats: formal methods for statistical analysis of neuroscience data. In addition to implementing neuroscience specific statistical techniques this could act as a binding to packages in JuliaStats.

[Marco-Congedo]

Hello,

I wrote specifically for EEG:

• FourierAnalysis.jl time series Fourier Analysis, in the frequency and time-frequency domain, including most measures of synchronization used in EEG/MEG

• PosDefManifold.jl, a package for manipulating covariance matrices

• PosDefManifoldML.jl, a package for classifying those covariance matrices

I am currentlly finalizing a package with all sort of diagonalization procedures, that is, the base algorithms to perform blind source separation (a kind of ICA), joint blind source separation, principal component analysis, maximum covariance analysis, canonical correlation analysis, the common spatial pattern and the common spatio-temporal pattern, a variant of the CSP for ERPs.

[Tokazama]

These look really nice. I'll take a look at them in more detail but it may make sense to have a sort of meta package for these.

[timholy]

I'd be happy to transfer https://github.com/timholy/AllenBrain.jl but I currently lack permissions to do so.

[Tokazama]

Thanks! Just sent an invite. This will probably be a good package to start figuring out exactly how we want to approach templates and ontologies.