presentation

[celiosantosjr]

User stories

• 1. Biochemist trying to figure out if their sequence was already observed and retrieve some info out of AMPSphere:
     • Example peptide: KRVKLLLKGYMRAIEINAALMYGYRPK
     • Search for similar peptides
     • Search for its family
• 2. Biochemist trying to produce or characterize one sequence from AMPSphere:
     • Example peptide: AMP10.000_010
     • Text search the access and retrieve the peptide information
     • Check features and peptide secondary structure
     • Check the genes
• 3. Ecologist trying to figure out the relationship between a certain bacteria in the microbiota of a given host:
     • Example host: Azolla sp. (a plant) and Nostoc azollae (a bacteria)
     • Browse filtering of the host and the original taxon, retrieve the list of AMPs
     • Entering the AMP access codes and check the location and habitat (guts, mouth, skin...)
• 4. Bioinformatician wanting to use AMPSphere as a resource:
     • Download the files
     • Retrieve the peptides and the genes
     • Retrieve families
• 5. Now imagine the case where the guy just wants to retrieve quick info and maybe export this to somewhere else, like your blog
     • use API
• 6. The same bioinformatician wants more info about a family of a certain AMP:
     • Example peptide: AMP10.000_010
     • Find the family, get the number of sequences, distribution in the world, the features, genes, environment, and hosts
     • Retrieve the HMM profile for the family and the logos for the sequence and the model

TODO:

+ implement a column and the filter in the browser tab for the AMP taxon origin (ex. E. coli)
+ implement file format and columns explanation in the download page 
+ add more explanation in the independent and conditional e-values in the HMMer search page (Family search)

[luispedro]

• i is very reasonable: many people will have sequences
• ii is possible if/once the resource is successful. But how did they even learn about AMP10.000_010?
• iii seems very forced. Why would AMPSphere be the best solution for this?
• iv feels under-specified. What does resource mean? Same comments applies tov (also it may not necessarily be a guy)
• vi: same comment as ii. Hopefully, we'll become successful enough that this is an actual use case

I am missing clicked on the link on twitter/manuscript/... and wants to know what this is all about.

Maybe also person has genome/MAG, wants to find all matches? is a common thing? Many people with genomes out there.