Getting started
Some advice for getting started with an hctsa analysis
🖥️ Talks and live demos

Youtube live stream of a tutorial session on feature-based time-series analysis tools at Computational Neuroscience Academy 2023, Krakow, Poland.

YouTube video walking through the tutorial on classifying EEG data, check out this slides for the full talk about analyzing neural dynamics using hctsa at CNS 2020 are here. July 2020.
💾 hctsa datasets, and example workflows
Sometimes it's easiest to quickly get going when you can follow on to some example code. Here are some repositories that allow you to do just this. There are a range of open datasets with pre-computed hctsa features, as well as some examples of hctsa workflows. You may also check the page of publications using hctsafor example publications that have used hctsa—some of these contain associated open code.

Seizure EEG. An overview tutorial on applying hctsa to a 5-class EEG dataset is in this GitHub repo (including the use of reduced feature set, catch22, within the hctsa framework). There is also a recording of the tutorial (final ~hour is a hands-on demo using this dataset) on YouTube.

Worms in a dish.
C. elegans movement speed data and associated analysis code.
You can try your hand at classifying different strains of the nematode worm C. elegans based on their time series of their movement speed. The repository, with links to pre-computed HCTSA.mat
files, is here.

Flies in a tube. Drosophila movement speed and associated analysis code. This repository allows you to skip the process of running an hctsa calculation (you can download pre-computed results), and get straight to following key analyses, including classifying day/night, or male/female.

1000 empirical time series. A collection of 1000 diverse empirical time series to test algorithms on.
If you have data to share and host, let me know and I'll add it to this list.
Last updated
Was this helpful?