Patent: Blockchain System with Nucleobase Sequencing as Proof of Work - Could this work for Gridcoin?
Intel corporation has applied for a patent describing a blockchain based system that uses the sequence of nucleotide bases within a nucleic acid to create a proof of work. [1]
Blockchain System with Nucleobase Sequencing as Proof of Work
Inventors: SMITH; NED M.; (Beaverton, OR) ; POORNACHANDRAN; RAJESH; (Portland, OR)
Abstract
A sequence mining platform (SMP) comprises a processor, at least one machine-accessible storage medium responsive to the processor, and a sequence manager in the machine-accessible storage medium. The sequence manager is configured to use processing resources to determine a sequence of nucleobases in a nucleic acid. The storage medium also comprises a blockchain manager to (a) collect transaction data for one or more transactions for a blockchain which requires a proof of work (POW) for each new block; and (b) include at least some of the transaction data in a new block for the blockchain. The storage medium also comprises a sequence mining module (SMM) to use the determined sequence of nucleobases from the sequence manager to create a POW for the new block. In one embodiment, the SMM enables an entity which controls the SMP to receive transaction rewards and sequencing rewards. Other embodiments are described and claimed.
Could something like this work for Gridcoin as a Proof of Work Mechanism?
It would be impractical at this point for us to implement a wet lab component for sequencing of unknown samples which I believe is the intention of the original Intel patent, since a lab would need to be set up containing next generation sequencing instruments (cost in the $100,000 - $1,000,000+ USD range) and qualified personnel would need to be hired. Although this type of thing could be quite viable if our market cap and price rises such that we could afford some type of facility for performing wet lab sequencing of samples and data storage.
We could potentially rely on existing labs to provide raw data for determination of nucleotide sequences. In fact, I think this could work well as a BOINC project. Researchers would be able to send raw sequencing data to a central BOINC server which then gets distributed into work units for BOINC crunchers to turn into usable DNA/RNA sequences. However there are some hurdles with this approach as well, since the major sequencing platforms (Illumina for example) have their own bioinformatics services (Basespace in Illumina's case) that have very fast turnaround. So the samples would need to from very large data sets (thousands or millions of human genomes, for example) such that Illumina's own compute clusters would be overwhelmed and would need to purchase or make use of additional computational resources. The other hurdle is that these next generation sequencing data sets are often in the 100 gig to several terabyte range so bandwidth would become a major issue. However, Once we had the infrastructure of a BOINC project/project server in place combined with raw data from sequencing instruments some type of proof of work mechanism based on nucleotide sequences may be viable.
References
[1] Smith, N. M., and Poornacandran, R. (2017). U.S. Patent Application 20170359374. Washington, DC: U.S. Patent and Trademark Office.
I love the idea of BOINC based Proof of Work to secure the blockchain. I don't know enough to know if this idea is usable on Gridcoin, but I think securing the blockchain with the research we are doing is an essential step going forward.
Don't forget to vote on the:
Gridcoin Proof Of Research Blockchain - Poll
I do think it would be usable but somewhat impractical if done the way it was described in the Intel patent, since we would need lab facilities etc.
Perhaps there is some way to match up work units from different gridcoiners and create some hash function from that, I dunno.