Design of DNA Storage Coding with Enhanced Constraints

Abstract

Traditional storage media have been gradually unable to meet the needs of data storage around the world, and one solution to this problem is DNA storage. However, it is easy to make errors in the subsequent sequencing reading process of DNA storage coding. To reduces error rates, a method to enhance the robustness of the DNA storage coding set is proposed. Firstly, to reduce the likelihood of secondary structure in DNA coding sets, a repeat tandem sequence constraint is proposed. An improved DTW distance constraint is proposed to address the issue that the traditional distance constraint cannot accurately evaluate non-specific hybridization between DNA sequences. Secondly, an algorithm that combines random opposition-based learning and eddy jump strategy with Aquila Optimizer (AO) is proposed in this paper, which is called ROEAO. Finally, the ROEAO algorithm is used to construct the coding sets with traditional constraints and enhanced constraints, respectively. The quality of the two coding sets is evaluated by the test of the number of issuing card structures and the temperature stability of melting; the data show that the coding set constructed with ROEAO under enhanced constraints can obtain a larger lower bound while improving the coding quality.

Study on DNA Storage Encoding Based IAOA under Innovation Constraints

“…However, as for the problem of the hairpin structure generation in DNA sequences, this problem can only be addressed to a certain extent by using the proposed constraints but cannot be completely solved. In [ 23 ], Li et al proposed a repeat tandem sequence constraint and improved the DTW distance constraint. The repeat tandem sequence constraint was used to address the problem of the secondary structure arising from successively repeated occurrences of coding sequences, and the improved DTW distance constraint was used to address the problem of inaccurate assessment of the overall similarity between sequences obtained under the traditional distance constraint.…”

“…The focus is on solving the problem of local base appearance. This study compared the lower bound data from the coding set with the results presented in [ 23 ] and found that the results presented in [ 23 ] show a good advantage in terms of finding the lower bounds of the coding sets of long sequences. However, for the lower bounds of the coding sets of shorter sequences, the results presented in this study are better than the results in [ 23 ] and [ 45 ], and some of the data presented in the study conducted by Li et al are inferior to those presented in [ 45 ].…”

“…This study compared the lower bound data from the coding set with the results presented in [ 23 ] and found that the results presented in [ 23 ] show a good advantage in terms of finding the lower bounds of the coding sets of long sequences. However, for the lower bounds of the coding sets of shorter sequences, the results presented in this study are better than the results in [ 23 ] and [ 45 ], and some of the data presented in the study conducted by Li et al are inferior to those presented in [ 45 ]. Because the constraints proposed in this paper and those proposed by Li et al [ 45 ] are different, the performance of the data in long and short sequences differs, which is also due to the different focus of algorithms, i.e., the IAOA proposed in this study and the ROEAO proposed by Li et al…”

“…In future studies, the characteristic that the constraint in [ 23 ] has a better limiting effect on long sequences could be combined with the proposed constraints to compensate for the shortcomings of this study. In addition, the algorithm in [ 23 ] has a good global search performance, and this advantage could be used to obtain more optimal solutions. Both of the mentioned aspects are crucial for constructing a large number of high-quality coding sets.…”

“…Zheng et al and Wu et al [ 21 , 22 ] developed methods for constructing well-performing DNA coding sets and introduced new constraints to construct more robust coding sets. In 2022, Li et al [ 23 ] proposed an efficient method for constructing DNA coding sets, which can effectively improve the quantity and quality of DNA coding sets. Ren et al [ 24 ] developed two highly reliable coding systems, named the RALR and the RABR, for four, six, and eight nucleotides.…”