<div>Hello Ugo Bruzadin Nunes,</div><div><br></div><div>By "Delorme et al., 2018" do you mean the paper "<span class="fontstyle0">Applying dimension reduction to EEG data by principal component analysis reduces the quality of its subsequent independent component decomposition</span>" published on NeuroImage?</div><div><br></div><div>In that paper, to achieve removing PCs accounting for 1% variance of the data, one needs to remove about 50 PCs out of the 71 EEG channels (see Figure 1C). If you are just removing 1 PC before ICA to make the data full-rank after re-referencing to average reference, I guess the negative effect of PCA will be much much smaller.</div><div><br></div><div>Best,</div><div><br></div><div>Yunhui<br></div><br><br><blockquote name="replyContent" class="ReferenceQuote" style="padding-left:5px;margin-left:5px;border-left:#b6b6b6 2px solid;margin-right:0">-----Original Messages-----<br>
<b>From:</b><span id="rc_from">"Ugo Bruzadin Nunes" <ugob@siu.edu></span><br>
<b>Sent Time:</b><span id="rc_senttime">2019-03-30 13:46:35 (Saturday)</span><br>
<b>To:</b> "eeglablist@sccn.ucsd.edu" <eeglablist@sccn.ucsd.edu><br>
<b>Cc:</b> <br>
<b>Subject:</b> [Eeglablist] Binica PCA error and PCA question<br><br>
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Dear SCCN team,</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="color: rgb(0, 0, 0); font-family: calibri, arial, helvetica, sans-serif; font-size: 12pt;"><br>
</span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="color: rgb(0, 0, 0); font-family: calibri, arial, helvetica, sans-serif; font-size: 12pt;">1. Me and the team at the INL in SIU-C have been trying to run a PCA using BINICA and have been getting the same error, no matter what computer, MATLAB,
or EEGLAB version we run it. We're running it on windows using the binica from the wiki page.</span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Binica's ICA runs ICA normally, and CUDAICA and RUNICA runs both ICA and PCA just fine. The binica I downloaded from the wiki, the CUDAICA for windows (in case you're curious, I found it in the comments of the original CUDAICA <a href="https://github.com/yhz-1995/cudaica_win" target="_blank">https://github.com/yhz-1995/cudaica_win</a> and
it works wonderfully).<span></span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
The code is this (or similar):" EEG = pop_runica(EEG,'icatype', 'binica', 'extended', 1, 'PCA', 25).</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
error print:</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
step 335 - lrate 0.000001, wchange 0.000000, angledelta 89.7 deg, 0 subgauss <br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span>Sorting components in descending order of mean projected variance ... <br>
</span>
<div>1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 <br>
</div>
<div>Permuting the activation wave forms ... <br>
</div>
<div>Storing weights in C:\MG/binica5727.wts <br>
</div>
<div>Storing sphering matrix in C:\MG/binica5727.sph <br>
</div>
<div> <br>
</div>
<span>ICA Version 1.3 (September 20, 1999) </span><br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span>Error using reshape<br>
</span>
<div>Size arguments must be real integers.<br>
</div>
<div><br>
</div>
<div>Error in floatread (line 158)<br>
</div>
<div>A = reshape(A,[Asize length(A)/prod(Asize)]);<br>
</div>
<div><br>
</div>
<div>Error in binica (line 326)<br>
</div>
<div> wts = floatread(weightsfile,[ncomps Inf],[],0);<br>
</div>
<div><br>
</div>
<div>Error in pop_runica (line 430)<br>
</div>
<span> [EEG.icaweights,EEG.icasphere] = binica( tmpdata, 'lrate', 0.001, g.options{:} );</span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
How can I fix this error?</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
2. I have read the many warnings from the SCCN team against running PCA on data to clean the data (Delorme et al., 2018), and I'm quite confused. Are we not supposed to run a "real PCA" or we're not supposed to run the "runica pca"? What alternatives do you
suggest to process the data? Should we run an n-1 ICA manual reduction to remove components but keep all the other ICs produced? Wouldn't an ICA of 128 on a 128 channels (in for example, a 10 minutes lenght file) be under-powered?</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
If my questions are repeated I apologize.</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Thanks a lot in advance,</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div id="signature">
<div></div>
<div></div>
<div id="divtagdefaultwrapper" dir="ltr" style="color:rgb(0,0,0); background-color:rgb(255,255,255); font-family:Calibri,Arial,Helvetica,sans-serif,EmojiFont,"Apple Color Emoji","Segoe UI Emoji",NotoColorEmoji,"Segoe UI Symbol","Android Emoji",EmojiSymbols,EmojiFont,"Apple Color Emoji","Segoe UI Emoji",NotoColorEmoji,"Segoe UI Symbol","Android Emoji",EmojiSymbols">
<div name="divtagdefaultwrapper" style="font-family:Calibri,Arial,Helvetica,sans-serif; margin:0px">
<span style="font-size:12pt">Ugo Bruzadin Nunes, <i>M.A.</i></span></div>
<div name="divtagdefaultwrapper" style="font-family:Calibri,Arial,Helvetica,sans-serif; margin:0px">
<div style="font-size:12pt"><span style="font-size:16px; background-color:rgb(255,255,255)"><i>PSYC 312 Instructor - Sensation and Perception<br>
</i></span></div>
<div><i>
<div style="display:inline!important"><font size="3">
<div style="display:inline!important">Brain and Cognitive Sciences Ph.D Program || Department of Psychology</div>
</font></div>
</i><br>
</div>
<div style="font-size:12pt"><span style="font-size:16px; background-color:rgb(255,255,255)"><i>
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:16px"><font size="3">
<div>Southern Illinois University - Carbondale</div>
</font></div>
</i></span>
<div style="font-size:16px; background-color:rgb(255,255,255)"><i></i></div>
</div>
</div>
</div>
</div>
</blockquote><br><br>