Alison Chaves

Foi a primeira vez que ouvi a pesquisadora se pronunciar sobre a polilaminina no #roda-viva. Fiquei assustado! Ela acha que está conduzindo o estudo do paraquedas.

Who are those guys? 4 retractions in a row today in ACS Omega!

I don't like those ultra-short gradient also, but we can't fight the data, as long as the experiment was properly done.

Well, that is it. Nyquist-Shannon theorem still holds. Despite beliefs, the data is the data.

This isn't validation. It just wast of time and reagents. Validation is when you find something significant and validate in an on-off system or suppressed gene dosage.

Agreed on the validation necessity. Strongly disagree on the method!

Yep, sometimes mad druids ask for WB as a form of validation for MS results.

A bit aggressive 😂

As far as I can say, they are consistent and work well. I often perform re-analysis of controlled datasets, and the changes are minimal.

I've seen several recent preprints using DIANN v1.9. DIANN is great and stable in several versions, except for v1.9. We tested at least 7 different versions in a publication 👉 pubs.acs.org/doi/10.1021/...
v1.9 will likely provide more IDs, but at the cost of precision.

If I may add something while this is a pre-print, it would be a re-analysis using any other version of DIA-NN, except v1.9.

Idea: A very rich person with Astral creates controlled mixtures and runs in conservative setup vs light speed Astral. If it's true, it's true.

This is by far the most important piece of advice I've read on social media lately.

Definitely not.

I once reviewed a manuscript that seemed ok, until I opened some raw files to do usual check and noticed that the TMT tags were incomplete in most of the spectra. Unfortunately, there are researchers that precludes access to data to force collaborations.

This kind of statement should be forbidden in science: "The data that support the findings of this study are available from the corresponding author upon reasonable request." Transparency must be encouraged. If there is no transparency, there is space for distrust.

1. Data available upon reasonable request." It is an MS journal 🤦‍♂️.
2. 50-100 mM TCEP is the stock concentration. Sample prep is 10x lower.
3. Overnight incubation with 100 mM 🤔.
4. With some effort, we can produce any artifacts.

Not yet. Positron has so many features and constant updates that make it hard to follow. I will try the databot at some point.

BTW, you will like to follow github.com/juliasilge

The main features I'm using are the "explain" and "fix" when I get some error in my code (lots of tricks with regex) and the autocomplete for repetitive code, like renaming things. If you prefer the chat, probably the providers have different performances.

Not happy with the choice on the DIANN version 1.9. To me, there is DIANN and DIANN 1.9!

QE and E480. Seems to be good

Finally, something with "cost-effective" in the title that hasn't used Astral.

MDPI again!

Absence of correlation between mRNA and protein levels. Probably some reviewer asked for validation of RNAseq results by qPCR. TMT analysis failed because the authors used MS2 resolution of 17,500 on a QE! The reviewer had no idea what they were evaluating.🥲
www.frontiersin.org/journals/imm...

PEA-based proteomics (Olink) is something we should pay more attention to. Apparently, people don't know how to properly normalize and do the stats in this kind of dataset. You should never live the statistic in the hands of core facilities.

I'm surprised that the authors declared that "Quality control measures and normalization of protein concentration were performed at the Firalis facility" and the reviewers were OK with it. It is probable that there is a normalization issue here.

I few the same about the cover arts. Real people are more creative in doing this kind of art.

That is true, but as far as I can tell they are enough to most tasks in our data analysis universe.

Switch to positron. It is way better than RStudio.