“The definition of insanity is doing the same thing over and over and expecting different results.”
Albert Einstein- PART 2
If you’ve been following along with the previous blog post, we explored what had been the constraints for the commercialization of aptamer-based diagnostics. Today, we’re picking up right where we left off explaining in more detail the FRELEX selection and the NEOMER library approach.
We developed our patented FRELEX selection method to enable the selection of aptamers against targets in their free state. We use a competing antisense on resin to partition aptamers that are bound to the target from aptamers that are unbound and hybridized to the antisense. This enables selection of targets in their entirely free state resulting in aptamers that work against targets as they exist in vivo.
We also invented the NEOMER library approach as a means to overcome the problems implicit in SELEX. SELEX selection is not reproducible. An aptamer library with a random region of 40 nucleotides has 1.2E24 possible sequences. If we had one copy of each sequence this would weigh 48 kg and is obviously impractical for selection. As such, SELEX is performed with a subset of possible sequences, generally around 1E15. This is such a small sampling of the possible number of sequences that it is actually unlikely that the same sequence would appear in two different aliquots of 1E15 sequences. The average copy number of each sequence when starting SELEX is one. This means that the vast majority of the best binding sequences are arbitrarily lost in the first round of selection, and it means that it is necessary to perform multiple rounds of reiterative selection in order to enrich sequences that bind. This has the potential to favour PCR bias. In all SELEX based selections there are enriched aptamer sequences that exhibit no binding at all.
In our NEOMER library approach we designed a library composed of 14 random nucleotides interspersed with fixed sequences. This makes sense because any given nucleotide cannot hybridize with other nucleotides within three bases, and thus a lot of structural variance potential is lost with random contiguous nucleotides. We used meta-analysis with thousands of possible templates and thousands of iterations of each template to define templates with high levels of structural diversity and complexity.
An aptamer library based on 14 random nucleotides has a total of 268,435,456 possible sequences. We apply an average of 10,000 copies of each of these sequences to every sample. The ability to start with 10,000 copies of each sequence means that we can drastically reduce the number of iterations of selection (even to only one round) and thus reduce potential for PCR bias. By using the same set of sequences in every application we have also made aptamer selection reproducible. We are able to apply the same library to the same target multiple times and extract average enrichment values and standard deviations of these values. This is the basis of science.
The key with the NEOMER library approach is that we can also apply the same starting library to counter targets. We then characterize performance of the best structures against the target in terms of selection against the counter targets. This is an in silico approach to mimicking immune tolerance, thus resulting in aptamers that will perform adequately in end use applications.
Much of the aptamer world appears to be continuing to apply SELEX to immobilized targets. Without a doubt this has resulted in many aptamers that perform extremely well against their targets, but to date it has not resulted in aptamers that are commercially successful in diagnostic applications.
NeoVentures is clearly leading the necessary reinvention of this science to drive commercial success. To learn more about innovative approach to aptamer development and our extensive use of machine learning to characterize binding structures please contact us at info@neoaptamers.com
Spanish 
English 
French