align.check with region-specific template alignment

Hello and thanks for this great software so far.

I am working with a custom V6-specific template alignment based on SILVA and have my sample aligned to it quite well. However, I do not understand the secondary structure map well enough in order to adjust it to the region of my template alignment, if that is possible. Is there a way to extract this map from ARB like you can extract sequences? I noticed that each line corresponds to a position in the full alignment and naively extracted it, but of course this did not work :frowning:

I appreciate your assistance in this forum.

The mothur secondary structure map is a list of numbers indicating what position that row forms a base pairing with. Modifying it is not for the faint of heart.

Ok, i did quick and dirty:

awk ‘{ if( NR >= 31188 && NR <= 33183 ){ p = $1 - 31187; if( $1 == 0 || p < 0 || p > 1996 ) print 0; else print p } }’ >

where the numbers given correspond to the 1-based positions for my target V6 region in the SILVA alignment. It left about 53 (106) paired bases in this region of the alignment. At least this map loads without error messages and I will check whether it works correctly, soon.

The align.check command cannot use paired positions with gaps in the alignment anyway, right, so I will not lose information?

Edit: I checked the specific map plus a left-shift and right-shift map to be sure that my calculated positions are correct. On the SILVA template alignment of the V6 region it gave me the following summed numbers of:
MAP pound dash plus equal loop tilde total
left-shift 439584 26112 18348 5644 2673366 27570 3190759
center 57484 139662 42676 72878 1217683 1458556 2989888
right-shift 651042 538 546 802 2657819 1156 3311912

From this I deduce that the V6-specific map was generated correctly.
And, of course the scoring includes gap-pairings but these would add up to the same constant for all the sequences in my sample, I think.

Any comments on the correctness of what I did?