Run python3 mmch.py AUGCUUC. the input sequence is the first argument.
For this I took a supersimplified aproach of the problem.
I am assuming that the Bonding graph is always two graphs separated, because no A or U base ever have an edge with any C or G, and vice versa.
Also, these two graphs have vertices between all their nodes, so basically all the alternate matchings between each of these graphs is the combinatory of the higher count of nucleotides of each type over the lower.
If you multiply these two, you should get the total different matching combinaitions.
Run python3 corr.py. the input file is in the same folder as input.fasta
I used the library Bio for reading fasta files, may require installation.
It's inside the API folder.
to run:
npm i
npm test (optional)
npm start
I didn't make many special assumptions, just tried to make it the easiest case possible.
Examples for the two endpoints:
curl -X GET 'http://localhost:3000/sequences?query=A&distance=20'
curl --data-raw '{"sequence":"ATCTC"}' -H "Content-Type: application/json" -X POST http://localhost:3000/sequences
If scaling up for many concurrent users, it would be useful to containerize (docker) and to use a non volatile storage (database).
The containerization makes it possible to use load balancing and autoscaling, to use more instances when the traffic gets higher.
Also, It may be needed to limit the access, by number of queries and require some permission to post.
If the posting is very heavy or has peaks, it might be needed to use async communication (queue) to receive the requests and save them with backpressure.
There are many other ways to improve the availability of the service or reduce it's cost, but they all depend on the specifics of the load.