CFM-ID’s latest version performs better in understudied metabolite classes
CliqueMS turns an old foe, in-source fragmentation, into a friend to build similarity networks for annotation
How the GNPS Dashboard improves collaboration from a web browser
How to use Cosine distance and Spec2Vec to compare spectra
AutoTuner is a new Bioconductor package that quickly and accurately computes preprocessing parameters for your MS2 metabolomics data
Developing metabolomics tools with doctors in mind
How to download, understand, clean, and analyze the GNPS JSON dataset
A unique machine learning strategy allows you to annotate each spectrum in an MS2 experiment with high accuracy on the chemical class level
How Feature-Based and Ion Identity Molecular Networking improve the accuracy of GNPS molecular networks
What software does our team use to keep knocking down the efficiency barriers?
A high level introduction to Dask
How we built Omigami: A scalable machine learning tool for metabolomic researchers
Leverage machine learning to detect poor quality integrations and save hours assessing peaks manually
Grokking the internals of Dask
Predict structure similarity from MS/MS spectra directly with a deep learning model.
What it’s like to be part of our team and why it might be different from what you expect
Properly tracking your machine learning experiments is easier than you think.
A new tool to smartly enrich molecular networks with metabolite annotations from different sources
Spec2Vec uses an unsupervised machine learning model to predict structural similarity from MS/MS mass spectra.
FEAST is the only standalone open-source feature store, but you have some other options too.
It’s Open Source, plays nicely with Kubernetes, and there’s a great community
A novel approach to feature annotation
Why we abandoned Kubeflow in our machine learning architecture
Insights from Nathan Wan
How to not overfit your models and get less false positives
Comparing machine learning platforms
Comparing data dashboarding tools and frameworks
Choosing a task orchestration tool
How can you process more data quicker?
How to deal with unseen data, optimise response times, and update models frequently.
How to parallelize and distribute your Python machine learning pipelines with Luigi, Docker, and Kubernetes
Leave your email to get our weekly newsletter.