Filling in missing values to make an array a set size.

I have written some code to detect retention times in the analysis of amino acids. Our current software is useless for this task. Using “for” and “if – else” loops, the code interrogates arrays and sequentially fills rows of data found at the right retention time into empty arrays full of NaN’s. Once done, if there are any rows of data that don’t belong to my target list, I remove those rows using “isnan”. My problem is that the reverse is also true; i.e. I sometimes have missing peaks. In order to export the data back to Excel for other colleagues I need to end up with a matrix which consistently has 22 “peak names” for each sample whether there is data in the row or not. I previously tried using “isempty” in the section of code that searches for the retention times. However, this didn’t work because I rely on an index assignment to fill my “NaN” template arrays an “isempty” returns an empty index. I can’t see how to directly index into the “NaN” template arrays during the retention time interrogation step because the “NaN” template arrays are shorter than the original data. I have tried using “intersect” comparing the array that I get after using “isnan” (which is shorter than the “NaN” template arrays) with a contrived array that contains the sample names and peak list that I want to end up with. But the index only gives perfect matches whereas I’m trying to find where there is a missing peak and insert a row with something like “not found” in it. I’ve also had a look at “ismember”, but the same thing applies. You get an index that is shorter than the “target, 22 x no. of samples” array and as such can’t be used.

ANY IDEAS?
BT 42936_-2339.dxf  BT 42936  CO2-1  1  126.4  22.4  753
BT 42936_-2339.dxf  BT 42936  CO2-2  2  225.9  22.4  753
BT 42936_-2339.dxf  BT 42936  CO2-3  3  325.4  22.4  754
BT 42936_-2339.dxf  BT 42936  CO2-5  6  614  22.6  755
BT 42936_-2339.dxf  BT 42936  Ala  7  702.9  29.7  252
BT 42936_-2339.dxf  BT 42936  Gly  8  741.3  41.8  332
BT 42936_-2339.dxf  BT 42936  Thr  9  799  23.2  265
BT 42936_-2339.dxf  BT 42936  Ser  10  822.8  20.9  347
BT 42936_-2339.dxf  BT 42936  Val  11  837.3  23.4  397
BT 42936_-2339.dxf  BT 42936  Leu  12  912.9  18.2  733
BT 42936_-2339.dxf  BT 42936  Ile  13  924.2  23.8  450
BT 42936_-2339.dxf  BT 42936  Nle  14  959.1  24.5  494
BT 42936_-2339.dxf  BT 42936  Pro  15  1051.7  21.5  481
BT 42936_-2339.dxf  BT 42936  Asp  17  1121.3  29.7  1216
BT 42936_-2339.dxf  BT 42936  Met  20  1167.9  21.1  225
BT 42936_-2339.dxf  BT 42936  Phe  23  1318  37.6  312
BT 42936_-2339.dxf  BT 42936  Tyr  24  1477  17.8  168
BT 42936_-2339.dxf  BT 42936  Lys  25  1505.4  44.7  923
BT 42936_-2339.dxf  BT 42936  CO2-6  29  1916.7  23.2  754
BT 42936_-2339.dxf  BT 42936  CO2-7  30  1953.9  22.6  758
BT 42936_-2340.dxf  BT 42936  CO2-1  1  126.4  22.4  754
BT 42936_-2340.dxf  BT 42936  CO2-2  2  225.9  22.4  754
BT 42936_-2340.dxf  BT 42936  CO2-3  3  325.4  22.4  754
BT 42936_-2340.dxf  BT 42936  CO2-4  5  574.3  22.8  759
BT 42936_-2340.dxf  BT 42936  CO2-5  6  614  22.6  755
BT 42936_-2340.dxf  BT 42936  Ala  7  702.2  28.2  260
BT 42936_-2340.dxf  BT 42936  Gly  8  740.5  44.3  336
BT 42936_-2340.dxf  BT 42936  Thr  9  798.2  23  267
BT 42936_-2340.dxf  BT 42936  Ser  10  822  21.1  327
BT 42936_-2340.dxf  BT 42936  Val  11  836.4  22.8  389
BT 42936_-2340.dxf  BT 42936  Leu  12  912.3  19  712
BT 42936_-2340.dxf  BT 42936  Ile  13  923.8  22.2  435
BT 42936_-2340.dxf  BT 42936  Nle  14  958.9  25.5  512
BT 42936_-2340.dxf  BT 42936  Pro  15  1051.3  21.3  482
BT 42936_-2340.dxf  BT 42936  Asp  16  1120.7  29.5  1188
BT 42936_-2340.dxf  BT 42936  Met  19  1167.3  26.3  221
BT 42936_-2340.dxf  BT 42936  Glu  20  1270.7  51.8  877
BT 42936_-2340.dxf  BT 42936  Phe  22  1317.5  36.6  279
BT 42936_-2340.dxf  BT 42936  Tyr  23  1476.2  20.1  160
BT 42936_-2340.dxf  BT 42936  Lys  24  1504.6  44.7  884
BT 42936_-2340.dxf  BT 42936  CO2-6  27  1915.5  22.2  756
BT 42936_-2340.dxf  BT 42936  CO2-7  28  1957.7  24.2  755

[Ashish Uthama]

Rhys, I cant seem to tie your explanation to the data posted (maybe its a 'domain' thing). Breaking down your processing into steps and posting data, with column headings, at each step might help us understand better.

[Fangjun Jiang]

Exactly! Well said, Ashish!