Find or invent a term for "Completely intersecting minus one"

I'm writing a paper that frequently references regions on a string, and these regions often intersect. I need to succinctly describe regions that almost completely intersect.

For example, given the string ACGT, suppose I have two regions described by [start,end] coordinates as follows:

R1=[0,2]="ACG"

R2=[1,3]="CGT"

Such that if these regions were laid side-by-side, we'd see that they intersect over the characters "CG":

``````ACG-
-CGT
``````

This particular situation arises a lot. I get two regions of equal length that intersect completely minus one. That is, I get two regions of length n that intersect by n-1 characters.

It would make my paper immensely clearer if I had a succinct way of stating this relationship. For a while I simply shoved "mis-" and "overlapped" together and said "mislapped". Thus instead of saying, "R1 and R2 intersect completely minus one," it would say "R1 and R2 are mislapped," or "R1 mislaps R2".

But "mislap" is clearly awful. Would the brilliant minds of English.StackExchange help me find or invent a better term?

• How about "plesiomorphic" (= nearly the same shape)? Or perhaps "plesiotopic" (=nearly in the same place)? Or "plesiogram" (I'm making them up now...) Sep 1 '13 at 22:45
• If they're disjoint, that's good enough. If their intersection is nonempty, you can say they intersect at "ACG", or their intersection is "ACG". If they both have the same length (`n`) you can say that they have an intersection value of from 0 (if they're disjoint) to `n` (if they're identical), with `n`-1 being the case outlined above. Sep 1 '13 at 23:05
• Are you specifically concerned with describing cases where one string overlaps another except for a single character? Supposing there was a word mmmph with precisely that meaning, so we could say ABC mmmphs AB. Would you want it to also be valid to say AB mmmphs ABC, or is that a different situation from your point of view? Realistically, how much use would such a word be? It would tend to imply that ABCDEFGHIJKLMNO mmmphs ABCDEFGHIJKLMNOP is of no more significance than AB mmmphs AQ. In the real world, the first case is probably important, and the second is just trivial. Sep 1 '13 at 23:29
• The fact that you are asking here on EL&U makes me think to suggest "-1". I.e., "R1 and R2 are -1ed", "R1 -1s R2" and so on. What about it?
– user19148
Sep 1 '13 at 23:33
• How about "nearly coincident" or "coincident less one"?? Sep 1 '13 at 23:46

Your situation reminds me of the Latinate word penultimate meaning “next to last,” from paene “almost” + ultimus “last.” You could combine the pen– prefix with another suitable Latinate word: penequivalent.

It looks like you are matching nucleotides, so I'm guessing this is related to bioinformatics. While I'm not conversant in the subject, I did come up with the term contig (as in "contiguous").

A contig (from contiguous) is a set of overlapping DNA segments that together represent a consensus region of DNA.

In the description of consensus sequence (link from consensus region), there is a discussion of notation for describing the quality of a match.

I would suggest you might try something like this:

"The strings R1 and R2 are contig(3,2)."

This would mean the strings are length 3 and have a contiguous match over 2 characters.

I don't know if this would create confusion with the existing use of "contig", but if you lay out this definition somewhere in an introduction, the reader should be able to follow.

If you are only interested in contiguous mismatches that are off by 1, then you might coin a shorthand term, like "R1 and R2 are contig-1".

• You guessed correctly. And yes, I've tried to avoid re-use of words that are already common bioinformatics parlance, but to define contig() as a function the way you do here is a fascinating idea I'll have to consider. Thanks! Sep 2 '13 at 1:12
• You're welcome. No doubt, if the reader comes across this form and overlooked your having defined it, they will look back to find what it means. You just have to make it easy to find. Sep 2 '13 at 1:26

The term for two sets that evaluate to be equal is coincident.

To describe your circumstance I believe the phrase coincident less one can be turned effectively.

This also easily generalizes to any small number of exceptions from coincidence.

• +1 for giving an answer that uses actual english words, and not newly-coined words or a domain specific function. ;) Sep 2 '13 at 8:59

Consider the following terms.
conjoined, “Joined together, as with conjoined twins, or in matrimony”
bonded, using senses of bond such as “A physical connection which binds, a band; often plural”, “An emotional link, connection or union”, and “Any constraining or cementing force or material”
friendly, reflecting that the two strings are quite alike
twined, using the “to become mutually involved” sense of twine
woven, using the “To unite by close connection or intermixture” sense of weave
mutual, adj., “Possessed in common”; you would use it as a noun, after properly defining it: “We call strings α and β mutual if...” • neighbors, in the sense “situated adjacently or nearby”
propinquinous (or even pinquinous), as formations based on propinquity, “Nearness or proximity; Affiliation or similarity”; or the Latin form propinquus, which has the following senses: “(of space) near, neighboring; (of time) near, at hand, not far off; (of appearance) resembling, similar, like, alike; (of a relationship) kindred, related”

I would advise not inventing a whole new word for this, it seems too narrow a concept to really demand one and instead refer to them as '(n-1)-overlaps'. I think this should be easily understood by your readers while remaining short enough to repeat often without becoming overly verbose. Alternatively if you really want to shorten further I imagine '(n-1)-laps' would be understood as it extends the form of 'k-mer' which is widely used in bioinformatics.

As you write it, it looks like you are talking of "substrings" (that is, pieces) of a unique given string. So "overlapped" or "mislapped" seems a particularly confusing teminology.

You might perhaps rephrase in terms of a "sliding window", or substring, or something similar. You could simply define it by the first index in the original string and the window width.

So that it would also make immediate sense for possible implementations.

• This is correct in terms of how I encode the information, but what I need is a term to succinctly express the particular situation described in the question. Sep 2 '13 at 1:18

I recommend

• discrete asymptote
• unit asymptote
• asymptotic unity
• discrete asymptotic unity

Asymptote
n.
A line whose distance to a given curve tends to zero. An asymptote may or may not intersect its associated curve.
[Ultimately from Greek asumpttos, not intersecting : a-, not; see a-1 + sumpttos, intersecting (from sumpiptein, sumpt-, to converge : sun-, syn- + piptein, to fall; see pet- in Indo-European roots).]