Premium
Semantic Retrieval in DNA‐Based Memories with Gibbs Energy Models
Author(s) -
Neel Andrew,
Garzon Max
Publication year - 2006
Publication title -
biotechnology progress
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.572
H-Index - 129
eISSN - 1520-6033
pISSN - 8756-7938
DOI - 10.1021/bp050141a
Subject(s) - latent semantic analysis , computer science , in silico , associative property , energy (signal processing) , gibbs free energy , artificial intelligence , theoretical computer science , mathematics , statistics , biology , physics , genetics , pure mathematics , gene , quantum mechanics
At least three types of associative memories based on DNA‐affinity have been proposed. Previously, we have quantified the quality of retrieval of genomic and abiotic information in simulation by comparison to state‐of‐the‐art symbolic methods available, such as LSA (Latent Semantic Analysis). Their performance is poor when the evaluation criterion for DNA‐affinity is a simple approximation of the Gibbs energy that governs duplex formation for retrievals. Here, we use a more realistic approximation of the Gibbs energy to improve semantic retrievals in DNA memories. Their performance is much closer to that of LSA, according to human expert ratings. With more realistic approximations of DNA affinity, performance is expected to improve for other, more adaptive associative memories with compaction in silico, and even more so with actual DNA molecules in vitro.