As pointed out by someone else on LinkedIn, the new anti obesity peptide recently published in Nature has the following MS/MS spectra 🤔🫣

Paper for reference: www.nature.com/articles/s41...

#proteomics #TeamMassSpec

After being asked to try 5% FDR to identify a protein.

#proteomics and #teammassspec things

TIL about "lasso peptides." #Peptides that make a literal lasso where the C-terminal chain loops through the ring.

Learn more here: is.gd/R1Vr8a
#science #proteins

How would they separate on reversed-phase material and ionize/fragment in a MS ? 🤔

Great slide by Ruedi Aeberold who kicks of #DAPSOC as the first speaker. #proteomics has been constantly outperforming "Moore's law" in regards to sensitivity (amount required) and throughput #science

Me on here 👏🏽👍🏽🔥🎉

Memes and vibes.

But no method is perfect, including this one. Each has its disadvantages and advantages and I think it important to be aware of the limitations. For example we thought we found a new Metformin target, but it is most likely an artifact of the method.

We validated a low affinity target (protein called Pirin) for Ibuprofen, which was found in different extracts.

This visualization is also a very nice way to look at shared drug targets from our dataset. A nice way to look at some complex data. Btw, the Cytoscape networks for all the drugs are downloadable (details below)

More details in the paper of course, but Nadya (co-author) used the resulting data to beautifully visualize drug-protein targets, I think it looks really cool.

Briefly, the strategy involves using the entire dataset to control for each compound (within each organ/cell extract experiment) to exclude drugs that are structurally similar or have known shared targets. We used the database drugcentral.org and the Tanimoto index for this.

I tried to take this up one more level and see if this technique could work on rat organs ex-vivo. Once the organs were collected, I tried to gently extract soluble proteins and use the same benchmark on different extracts. It works well for most extracts.

We used an optimized version of the PISA approach where we forgo the use of TMT reagents, use a label-free DIA MS method, optimize heating temperatures + filter plates. It works well in our Staurosporine ("positive control") benchmark. Look at this beautiful volcano plot 🌋

TPP can measure melting point of 1000's of proteins from a complex mixture (ie. cells, organ extracts), which can change if a protein is targeted by a drug. There are variations of this principal that integrate melting curve such as proteome integral solubility alteration (PISA)

Our totally unbiased view is: if you are trying to determine protein targets, start with measuring the actual proteins (ie. mass spectrometry [MS] based proteomics). One the most prominent method to this is based on Thermal Proteome Profiling (or its many variations such as PISA)

And there are MANY methods that attempt find these unknown targets, ultimately it will be a trade off between speed, specificity, and the proteome depth. There is NO perfect method, and each will have its advantages and disadvantages.

Ok, going to try to repost the thread on here as well :)
---------
I'm happy to report the publication of our latest #proteomics paper has been published in Nature Communications
. Thanks to the reviewers and editors who helped us improve the manuscript!
rdcu.be/dZLbk
A small 🧵below

Why not? #science #Teammassspec