Lightnews — Scholar-powered news

unitsaq.bsky.social @unitsaq.bsky.social · 3d

SPOT-1000 and others from divbio.com/labels/cryo-... compatible with laser printers. Once the excel spreadsheet is set up it's easy to populate the sample ids

Diversified Biotech | Laboratory Labeling Solutions & Planet-Safe® Labware

Developers and manufactures of specialized labeling solutions such as cryogenic labels (Cryo-Tags® & Cryo-Babies®), freezer labels (Tough-Tags®), application-specific labels for various labware, as we...

Divbio.com

1

unitsaq.bsky.social @unitsaq.bsky.social · 20d

I sometimes use Waters PrepMass column calculator for backpressures, volumes, flows, loads, and translating between column dimensions but it doesn't scale down to nanoflow. So I cloned and modified with LLM in a few shots: unitsaq.shinyapps.io/columncalcul...
Original:
www.waters.com/webassets/ot...

PrepMass Column Comparator (Scaled to μL/min and μg)

unitsaq.shinyapps.io

unitsaq.bsky.social @unitsaq.bsky.social · 21d

Afaiu current LLMs can't imagine completely new concepts but they can create novelty via concept shuffling / information transfer. Analogous to translating deja vu to English.

unitsaq.bsky.social @unitsaq.bsky.social · 28d

When I first learned about spectral counting as a way of quantifying proteins I was stunned at how "qualitative" it seemed. Now I think it served it's time--how else are you supposed to "quantify" proteins from hyperfractions across dozens of labs' LC-MS systems eg the Human Plasma Proteome Project.

unitsaq.bsky.social @unitsaq.bsky.social · Sep 5

This is gold

1

unitsaq.bsky.social @unitsaq.bsky.social · Aug 28

My treatment significantly modulates gene expression machinery!

1 1

unitsaq.bsky.social @unitsaq.bsky.social · Aug 28

Remember to say please & thank you (& tip $ if you don't mind potential future payback risk 😜)

Definitely some new LLM-powered bioinformatic tools to be developed with careful engineering e.g. interpreting top hits based on the human knowledgebase (analogous to backgrounded pathway analysis)

1

unitsaq.bsky.social @unitsaq.bsky.social · Aug 14

It's ironic that people often use/suggest fetus-derived cells eg HEK293 as a ethical substitute for HeLa.

unitsaq.bsky.social @unitsaq.bsky.social · Aug 14

Nice work, demonstrating determination of omega positions in lipid acyls using only RT prediction! www.nature.com/articles/s41...

Computationally unmasking each fatty acyl C=C position in complex lipids by routine LC-MS/MS lipidomics - Nature Communications

Physiologically relevant omega-positions of double bonds in fatty acyls in complex lipids can now only be identified with specialized instrumentation. Here, the authors present a computational approac...

www.nature.com

3 4

unitsaq.bsky.social @unitsaq.bsky.social · Aug 6

#TeamMassSpec Bench Tip: get a 5-kg scale with subgram accuracy for making >200 mL solutions.

Faster and more accurate than a graduated cylinder.

1) Measure out solute mass
2) Calculate mass of solvent wrt solution and solvent density (in a reusable spreadsheet)
3) Add solvent mass, done!

unitsaq.bsky.social @unitsaq.bsky.social · Aug 4

If you haven't increased the DDA tolerance to 50 mz window and set it to go after off inclusion list ions while waiting, then the instrument may be waiting around for the inclusion list mzs.

1 1

unitsaq.bsky.social @unitsaq.bsky.social · Jul 28

Would be interesting to see the effects of constant [trypsin] without normalizing [protein]. May I (or someone else) can compare?

unitsaq.bsky.social @unitsaq.bsky.social · Jul 28

Yes but that conclusion may lead to the misinterpreted addition of the same [trypsin] at wildly different [protein] which I can't imagine will yield the same proteomics profiles due to concentration-dependent effects. Still believe [protein] needs to normalize pre-digestion AFAIU.

1 1

unitsaq.bsky.social @unitsaq.bsky.social · Jul 28

Distilled rule of thumb for optimal Protein:Trypsin ratio vs Total Protein concentration in bottom-up #proteomics:

Optimal protein:trypsin ≈ 0.1 * protein concentration (ng/uL)

Aligns well with standard guidelines e.g. 50:1 at ~500 ng/uL

Data from Table 5 in www.sciencedirect.com/science/arti...

1 1 8

unitsaq.bsky.social @unitsaq.bsky.social · Jul 24

Distilled rule of thumb for optimal Cosine cutoffs at various n matching MS2 fragments in #metabolomics ID:

Cosine > 0.90 - 0.05*MMP

Aligns well with ~80% at 3 matched fragments I learned before.

From Fig4a here www.nature.com/articles/s41...

1 2

unitsaq.bsky.social @unitsaq.bsky.social · Jul 19

Are these all these sentimentals just about the affinity platforms, or LC-MS too? I thought large plasma datasets would be helpful for filtering noise a la GWAS. AFAIR the UK Biobank plasma disease studies were anti-hopium with mostly minor improvements of Dx accuracy vs traditional biomarkers.

2

unitsaq.bsky.social @unitsaq.bsky.social · Jul 7

Do the splitting ones have any special characteristics eg reduced solubility or amphipathic nature? How big is the first vs second peak? Speculating, maybe the first elution point doesn't have enough time/flow to elute everything and the second elution point cleans up the rest?

1 1

unitsaq.bsky.social @unitsaq.bsky.social · Jul 1

I was never able to get two scan boxcar-DDA to ID more proteins than a vanilla top 15 DDA on my QEP. Probably could do better with maxIT > 5 ms for the 10x 50mz msx tsim MS1 scan but then it would start eating MS2 time so I never ventured.

1

unitsaq.bsky.social @unitsaq.bsky.social · Jun 27

Supposedly Edison and others held objects in hand that would drop as they fell asleep, wakimg them in a creative cognitive interphase, aka using a "hypnagogia catcher"

unitsaq.bsky.social @unitsaq.bsky.social · May 13

Diabetic fasting glucose is 1.3x

unitsaq.bsky.social @unitsaq.bsky.social · May 8

doi.org/10.1101/2025... unbroken link

An adaptive, continuous-learning framework for clinical decision-making from proteome-wide biofluid data

Mass spectrometry (MS)-based proteomics provides deep molecular insights from patient samples, but clinical use has been limited by missing values, static biomarker panels, and the need for targeted a...

doi.org

unitsaq.bsky.social @unitsaq.bsky.social · Apr 30

Careful interpretation key before and after datasets collected. Always risk unforeseen unmeasured confounders even with the extensive planning.

unitsaq.bsky.social @unitsaq.bsky.social · Apr 30

Well-designed experiments are ideal, but looking at public repositories, they are not as common as one would hope.

1

unitsaq.bsky.social @unitsaq.bsky.social · Apr 30

With sufficient data breadth/size, major unmeasured confounders could theoretically be inferred and accounted for by their children/proxies eg red blood cells contamination by hemoglobin, socioeconomic status by specific exposure patterns, DMSO by previously observed effects.

1

unitsaq.bsky.social @unitsaq.bsky.social · Apr 22

Nice to see a systematic characterization of flow vs performance. Looks like if you can give up the 10x boost in sensitivity, microflow is the same or better vs nano which was also shown by Bian et al pubmed.ncbi.nlm.nih.gov/31919466/#&g...

Robust, reproducible and quantitative analysis of thousands of proteomes by micro-flow LC-MS/MS - PubMed

Nano-flow liquid chromatography tandem mass spectrometry (nano-flow LC-MS/MS) is the mainstay in proteome research because of its excellent sensitivity but often comes at the expense of robustness. He...

pubmed.ncbi.nlm.nih.gov

1