A Table of Bounds on Optimal Fixed-Length Quaternary Edit-Metric Codes

 n\d 1 2 3 4 5 6 7 8 9 10 11 12 13
 1 4 1 1 1 1 1 1 1 1 1 1 1 1
 2 16 4 1 1 1 1 1 1 1 1 1 1 1
 3 64 16 4 1 1 1 1 1 1 1 1 1 1
 4 256 64 16 4 1  1  1  1  1  1 1 1 1
 5 1024 256 44 11 4  1  1  1  1  1 1 1 1
 6 4096 1024  114-176 28-32 8 4  1  1  1  1 1 1 1
 7 16384 4096 356-614 65-128 18-24 8 4 1  1  1 1 1 1
 8 65536 16384 1132-2340 176-512 38-96 14-32 5 4  1  1 1 1 1
 9 262144 65536 3451-9360 495-2048 97-384 27-128 11-20 5 4  1 1 1 1
10 4^10 262144 11743-30417 1463-8192 249-1536 60-496

20-80

10-16 5 4 1 1 1
11 4^11 4^10 40604 4574 676 139 41 16 7 4 4 1 1
12 4^12 4^11 40604 4574 1894 338 79 30 13 6 4 4 1
13 4^13 4^12 40604 7744 5517 879 188 54 17 11 5 4 1
14 4^14 4^13 40604 15206 5517 879 188 54 27 13 7 5 4
15 4^15 4^14 40604 17780 6419 880 196 61 45 19 10 6 4
16 4^16 4^15 40604 18308 9220 1784 333 196 83 32 16 8 6
17 4^17 4^16 40604 18308 9220 1784 333 322 143 57 24 13 8

The above table shows E4(n,d), the number of codewords in an optimal quaternary code of fixed length n and minimum edit distance d. For n ≤ 10, exact values are given where known, with upper and lower bounds provided when an exact value is not known. For n > 10, the value shown is the lower bound on E4(n,d).

The earliest version of this table appeared in the technical report: S.K.Houghten, D.Ashlock and J.Lenarz, "Bounds on Optimal Edit-Metric Codes"and was ultimately published as "Construction of Optimal Edit Metric Codes", Proceedings of the 2006 IEEE Workshop on Information Theory (ITW 2006), 259-263.

Improvements since that time:

  1. E4(5,3) ≥ 44 and E4(5,3) ≤ 46 (D.Ashlock & S.Houghten, July 2005)
  2. E4(6,4) ≤ 32 and thus E4(7,4) ≤ 128, E4(8,4) ≤ 512, E4(9,4) ≤ 2048 and E4(10,4) ≤ 8192 (D.Ashlock & S.Houghten, July 2005)
  3. E4(7,5) ≤ 28 and thus E4(8,5) ≤ 112, E4(9,5) ≤ 448 and E4(10,5) ≤ 1792 (D.Ashlock & S.Houghten, July 2005)
  4. E4(5,3) = 44 and thus E4(6,3) ≤ 176 (S.Houghten, August 2005)
  5. E4(7,5) ≤ 24 and thus E4(8,5) ≤ 96, E4(9,5) ≤ 384 and E4(10,5) ≤ 1536 (S.Houghten, August 2006)
  6. E4(6,3) ≥ 108 (D.Ashlock and S.Houghten, October 2008)
  7. Table expanded to n=16 (R.Flack & S.Houghten, 2008)
  8. E4(10,9) ≥ 5, E4(11,7) ≥ 34 (and thus E4(12,7) ≥ 34) , E4(11,8) ≥ 14 (and thus E4(12,8) ≥ 14), and E4(11,9) ≥ 7 (and thus E4(12,9) ≥ 7)(D.Ashlock & S.Houghten, March 2010)
  9. E4(6,3) ≥ 111, E4(7,3) ≥ 343, E4(8,3) ≥ 1082, E4(7,4) ≥ 65, E4(8,4) ≥ 173 (D.Ashlock, S.Houghten, J.A.Brown & J.Orth, September 2010)
  10. E4(9,4) ≥ 495, E4(9,5) ≥ 97, E4(9,6) ≥ 27, E4(9,8) ≥ 5, E4(10,6) ≥ 59, E4(10,7) ≥ 20, E4(11,5) ≥ 676, E4(11,6) ≥ 133 (D.Ashlock, S.Houghten, J.A.Brown & J.Orth, September 2010)
  11. E4(11,3) ≥ 40604, E4(11,4) ≥ 4574 (and thus E4(12,4) ≥ 4574), E4(12,5) ≥ 1894, E4(13,5) ≥ 5517 (and thus E4(14,5) ≥ 5517), E4(11,6) ≥ 133, E4(12,6) ≥ 338, E4(13,6) ≥ 879 (and thus E4(14,6) ≥ 879), E4(12,7) ≥ 79, E4(13,7) ≥ 188 (and thus E4(14,7) ≥ 188), E4(12,8) ≥ 28, E4(13,8) ≥ 54 (and thus E4(14,8) ≥ 54) and E4(12,9) ≥ 12 (and thus E4(13,9) ≥ 12) (D.Ashlock, S.Houghten, J.A.Brown & J.Orth, September 2010)
  12. E4(6,3) ≥ 114, E4(7,3) ≥ 356, E4(8,3) ≥ 1132, E4(8,4) ≥ 176, E4(10,5) ≥ 249, E4(10,6) ≥ 60, E4(11,6) ≥ 139, E4(11,7) ≥ 41, E4(11,8) ≥ 16, E4(12,8) ≥ 30, E4(12,9) ≥13 (and thus E4(13,9) ≥ 13): see D.Ashlock, S.K.Houghten, J.A.Brown and J.Orth, On the synthesis of DNA error correcting codes, BioSystems 110 (2012), 1-8.
  13. E4(12,10) ≥ 6, E4(13,9) ≥ 17, E4(13,11) ≥ 5, E4(14,9) ≥ 27, E4(14,10) ≥ 13, E4(14,11) ≥ 7, E4(14,12) ≥ 5, E4(15,9) ≥ 45, E4(15,10) ≥ 19, E4(15,11) ≥ 10, E4(15,12) ≥6, E4(16,9) ≥ 83, E4(16,10) ≥ 32, E4(16,11) ≥ 15, E4(16,12) ≥ 8, and addition of values for E4(17,d): see D.Ashlock and S.Houghten, Lexicode Crossover for Embeddable Biomarkers, IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, 2015.
  14. E4(16,8) ≥ 196, E4(16,11) ≥ 16, E4(17,8) ≥ 322, E4(17,9) ≥ 143, E4(17,10) ≥ 57, E4(17,11) ≥ 24, E4(17,12) ≥ 13, and addition of values for E4(n,13), D. Ashlock and S. Houghten, Hybridization and Ring Optimization for Larger Sets of Embeddable Biomarkers, 2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, 2017.

If you have improvements, comments, etc, please e-mail shoughten@brocku.ca

This page last modified 13th November, 2023
© Copyright Sheridan Houghten, 2005-2023