TESA example data
The TMS-EEG signal analyser (TESA) is an open source extension for EEGLAB that includes functions necessary for cleaning and analysing TMS-EEG data. Both EEGLAB and TESA run in Matlab (r2015b or later). The attached files are example data files which can be used with TESA.
To download TESA, visit here:
To read the TESA user manual, visit here:
WARNING: file size = 1.1 GB. A raw data set for trialling TESA. Load the data file in to EEGLAB using the existing EEGLAB data set functions. Note that both the .fdt and .set files are required.
File size = 340 MB. A partially processed data file of smaller size corresponding to step 8 of the analysis pipeline in the TESA user manual. Channel locations were loaded, unused electrodes removed, bad electrodes removed, epoched (-1000 to 1000 ms) and demeaned (baseline correct -1000 to 1000). Load the data file in to EEGLAB using the existing EEGLAB data set functions. Note that both the .fdt and .set files are required.
File size = 69 MB. A further processed data file even smaller in size corresponding to step 11 of the analysis pipeline in the TESA user manual. In addition to the above steps, data around the TMS pulse artifact was removed (-2 to 10 ms), replaced using linear interpolation, and downsampled to 1,000 Hz. Load the data file in to EEGLAB using the existing EEGLAB data set functions. Note that both the .fdt and .set files are required.
Example data info:
Monophasic TMS pulses (current flow = posterior-anterior in brain) were given through a figure-of-eight coil (external diameter = 90 mm) connected to a Magstim 2002 unit (Magstim company, UK). 150 TMS pulses were delivered over the left superior parietal cortex (MNI coordinates: -20, -65, 65) at a rate of 0.2 Hz ± 25% jitter. TMS coil position was determined using frameless stereotaxic neuronavigation (Localite TMS Navigator, Localite, Germany) and intensity was set at resting motor threshold of the first dorsal interosseous muscle (68% maximum stimulator output). EEG was recorded from 62 TMS-specialised, c-ring slit electrodes (EASYCAP, Germany) using a TMS-compatible EEG amplifier (BrainAmp DC, BrainProducts GmbH, Germany). Data from all channels were referenced to the FCz electrode online with the AFz electrode serving as the common ground. EEG signals were digitised at 5 kHz (filtering: DC-1000 Hz) and EEG electrode impedance was kept below 5 kΩ.