# Struggling with ICA

**URL:** <https://mne.discourse.group/t/struggling-with-ica/8000>\
**Category:** Support & Discussions\
**Tags:** eeg\
**Created:** [December 13, 2023, 6:35pm UTC](https://mne.discourse.group/t/struggling-with-ica/8000 "2023-12-13T18:35:31Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![vxnuaj](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/vxnuaj/32/3267_2.png) [@vxnuaj](https://mne.discourse.group/u/vxnuaj)\
**Post date:** [December 13, 2023, 6:35pm UTC](https://mne.discourse.group/t/struggling-with-ica/8000/1 "2023-12-13T18:35:31Z")

</div>

Hey everyone!

I’m currently using ICA to filter our any artifacts from my dataset.

Here’s the issue:

Prior, I applied bandpass/stop filters with a Notch filter to minimize power line interference at 50Hz. When I look at my PSD plot, there is no spike at 50 Hz

Yet when I conduct ICA and look at individual components, I’m seeing a singular spike at 50 Hz.

Why is this the case?

I’m linking a Jupyter notebook below for inspection if you’re willing to help out!

I’d highly value any advice!

> <https://github.com/vxnuaj/mneprocess/blob/main/MNE-EEG-PROCESSING/Replicate2.ipynb>

---

<div class="post-metadata">

**Author:** ![drammock](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/drammock/32/4_2.png) [@drammock](https://mne.discourse.group/u/drammock)\
**Post date:** [December 13, 2023, 9:32pm UTC](https://mne.discourse.group/t/struggling-with-ica/8000/2 "2023-12-13T21:32:36Z")

</div>

you are doing `ica.plot_properties(raw, ...)` in cell 19. So even though ICA was computed based on filtered data, the plotting function is _applying_ the ICA to the _unfiltered_ data, which is why you see the power line spike.

---

<div class="post-metadata">

**Author:** ![vxnuaj](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/vxnuaj/32/3267_2.png) [@vxnuaj](https://mne.discourse.group/u/vxnuaj)\
**Post date:** [December 14, 2023, 12:35am UTC](https://mne.discourse.group/t/struggling-with-ica/8000/3 "2023-12-14T00:35:24Z")

</div>

oml, that as to be the most obvious mistake – right under my eyes yet I couldn’t see it.

Thanks for your help!