Ever get lost searching through genetic literature? I've built a new interface for bioRxiv+medRxiv to directly browse preprint findings on the genome!To begin I downloaded 7TB's of bioRxiv+medRxiv preprints and extracted mentions of genetic variant ID's or positions. Documented in "How to download bioRxiv on a budget": http://sitlabs.org/writing/biorxiv.htmlThis project started after I was...
https://www.barneyhill.com/pages/genomie/
I built a natural language interface for semantically searching across FutureHouse's WikiCrow generated gene summaries embedded in a genome browser.As a PhD student working with non-coding variants, I found it challenging to quickly interpret genetic associations using traditional genome browsers. Genomie helps prioritize genes that could be regulated by nearby non-coding associations by...
Hopefully I'll be able to add more datasets in the coming weeks - if you know any do shout!
https://www.barneyhill.com/pages/fieldstar/
This project builds upon cosine.club - an amazing music similarity search engine posted the other day (https://news.ycombinator.com/item?id=40072062) which works by comparing audio embeddings between songs with this new model from essentia/upf https://essentia.upf.edu/models.html#discogs-effnet.I wondered what these embeddings would look like when visualised (using UMAP). Would genre "clusters"...
Was brilliant, a lot of groups focusing on the use of ML to characterise the "unexplainable" in sound synthesis.
We ended up submitting a tool for interacting directly with Ableton using LLM agents after becoming disenchanted with text2audio models, wrote about it here - https://montyanderson.net/writing/synthesis
We just published a blog today discussing this - https://montyanderson.net/writing/synthesis
It works by displaying all sets containing the selected track. You can then click through white nodes to traverse the network or search for a specific song/artist. All the data is downloaded to the client and traversal is done locally so the page is static.