Matterworks

We found a previously unreported biomarker in one of the best annotated lipidomics datasets in the public domain! The lipid was PC 32:3. It's been a fixture of commercial targeted lipidomics panels for years, with a steady accumulation of evidence tying it to insulin resistance and metabolic disease. When we ran the raw MS files through our new de novo lipid ID model, PC 32:3 came back with clear differential abundance between lean and obese groups. On the same raw files, our model reproduced every lipid class the original study reported and added 222 new species on top. Today we're launching that model in Pyxis, as the first fully de novo lipid identification model on the market! The dataset is loaded in Pyxis as a demo. Sign up and take a look ⬇️

Nice work by @GabrielAsher02 and the team from @matterworks_bio! alphaxiv.org/abs/2510.26715

Two public studies are pre-loaded for exploration (Antipsychotic Overdose, Cystic Fibrosis), and you can upload your own. Try it: app.matterworks.ai

We just trained LSM-MS2, a large spectral foundation model built on millions of mass spectra. It learns chemical semantics directly from raw MS/MS, so the ~80% of “unknown” features can finally be used for biological interpretation and phenotype stratification with no compound IDs required. For traditional annotation, it also exceeds SOTA: • +30% isomer discrimination • +42% correct calls in complex matrices • Accurate even at trace concentrations

We're excited to be at #CHIBioprocessingSummit next month, showcasing our AI platform that unlocks deeper biological insights for biotech discovery. @BaljitKUbhi, @hernanmomel, James Moore, Adi Kulkarni, and Jack Howland will be onsite. Will you be there? #Matterworks #AI #Boston

ASMS 2025 was electric, and Matterworks showed up. ⚡ From LSM demos 🔬 to deep scientific conversations 🧠, we have proven that machine intelligence 🤖 isn’t an add-on. It is the foundation of modern omics. #ASMS2025 #MachineIntelligence #Omics #PredictiveBiology

⚡Today's Matterworks posters at #ASMS2025 ThP 455: Generalizing Pyxis across HILIC- and RP-compatible analytes ThP 086 (with @corteva): Pyxis’ application to water treatment metabolic degradants ThP 352 (with @bruker): Implementing Pyxis on an IM-ToF-MS hubs.la/Q03q11NX0

⚡Visit our' Wednesday poster at #ASMS2025 WP 375 (with Metalytics): Metabolic flux analysis for improvements in biomanufacturing hubs.la/Q03q119x0

⚡Our posters today at #ASMS2025 TP067: Phenotype prediction in precision oncology and the use of Pyxis in: TP064 (@dsmfirmenich): Microalgal fermentation TP239 (@310ai__) Incretin-mediated metabolic reprogramming TP582 Mitochondrial dysfunction hubs.la/Q03py-x00

⚡Today's Matterworks posters at #ASMS2025 MP 626 (with J&J): Elucidating CHO Cell Metabolism with an ML-Based Platform MP 410 (with Thermo Fisher): Using LSMs for complex samples MP 631: Broadened coverage for rapid absolute quantitation of metabolites hubs.la/Q03pvtF90

Don't miss the #ASMS2025 @thermosci Users Meeting today, where Dr. Jennifer M. Campbell will present our progress in self-supervised machine intelligence for #predictivebiology. Registration: hubs.la/Q03py_rx0 Metabolomics breakout: 10:15am Baltimore Marriott Waterfront

Matterworks would like to thank our #ASMS2025 poster collaborators: @JNJNews, @dsmfirmenich, @corteva, @bruker, @310ai__, @thermosci, and Metalytics. 🔎 For a full list of our posters, visit (hubs.la/Q03pxTYV0) Thank you to all the organizers for an excellent program.

Join the #ASMS2025 @thermosci Users Meeting to hear Dr. Jennifer M. Campbell discuss our progress in self-supervised machine intelligence for #predictivebiology. Registration: hubs.la/Q03pv6r-0 Metabolomics breakout: 10:15am Baltimore Marriott Waterfront

Submitting #ASMS abstracts at the beginning of February reminded us of the great work our team did at last year's ASMS meeting. The video summarizes a poster in which the performance of Pyxis was compared against traditional isotopologue analysis. hubs.la/Q03byDbM0

Did you miss our webinar? Or would you like a recap? Download our editorial summarizing the discussion of Pyxis for quantitative analysis in disease research, featuring Assoc. Prof. Naama Kanarek and Dr. Jennifer Campbell. hubs.la/Q037qFtJ0

We're excited to attend the inaugural Predictive Phenomics conference hosted by Pacific Northwest National Laboratory next month. We'll follow up shortly on the work we'll showcase.

It's the final day of #Pittcon! Find the Matterworks team at booth 1853 today from 10 AM to 3 PM.

Today at #Pittcon Matterworks CSO Jennifer Campbell will be presenting on how we're leveraging AI to identify and quantify molecules from raw LC-MS data. The talk will take place at 4:20 PM in Room 210A.

#Pittcon is taking place in Boston for the first time & we're excited to participate! Our technology is being presented in the main program, and we'll be on the exhibition floor at booth 1853. Come and hear about our AI-powered technology for raw LC-MS data interpretation.