Sign in

Alison Chaves

@alisonchaves.bsky.social
218 followers 263 following 490 posts

Doing proteomics until I still can. github.com/41ison

PostsRepliesMedia
Alison Chaves @alisonchaves.bsky.social · 03/10/2026
Yep pubs.acs.org/jprobs/artic...
020
Alison Chaves @alisonchaves.bsky.social · 02/10/2026
Let's wait for the presidential election result here to find out if we're happy.
010
Alison Chaves @alisonchaves.bsky.social · 02/10/2026
Not sure whether you are happy, but I'm 😂
120
Alison Chaves @alisonchaves.bsky.social · 30/09/2026
Also, these days IF dons't mean anything. Several questionable journals have high IF. When we apply to a position, often we are asked why we don't have a publication in bla, bla, bla journal.
000
Alison Chaves @alisonchaves.bsky.social · 30/09/2026
Something I'm always bothered about. In academia there people say this about negative results. Journals usually don't like it! The reviewers don't like it as well.
100
Alison Chaves @alisonchaves.bsky.social · 26/09/2026
Interesting discussion. I'm going to implement TMM as an option for normalization in my R tool.
000
Alison Chaves @alisonchaves.bsky.social · 17/09/2026
Some companies selling Abs are visionaries! They are selling the Abs using AI for the results.
000
Alison Chaves @alisonchaves.bsky.social · 16/09/2026
Weird volcano plot! Not the asymmetry, but the distribution pattern. One we get there, not yet. Tesorai is the only credible thing I've seen like this.
110
Alison Chaves @alisonchaves.bsky.social · 10/09/2026
So, are you telling me AI is stealing a scientist's work that the general audience doesn't know and presenting it as original pieces? 🤔AI is finally similar to humans, impressive!
000
Alison Chaves @alisonchaves.bsky.social · 01/09/2026
I'm impressed how using the prompt "use an appropriate statistical method" for the Olink example returned a credible representation. It is better than the z-score made by humans publishing in nature.
100
Alison Chaves @alisonchaves.bsky.social · 27/08/2026
Bonsai reconstructs tree representations for distortion-free visualization and exploration of high-dimensional data www.nature.com/articles/s41...
nature.com
000
Alison Chaves @alisonchaves.bsky.social · 20/08/2026
Congrats. It is a very well done work.
100
Alison Chaves @alisonchaves.bsky.social · 20/08/2026
Well, the irony is that if we want to read this column:
010
Alison Chaves @alisonchaves.bsky.social · 19/08/2026
500k-1M resolution for resolving 19 mDa is a bit exaggerated. But I got the point. You really don't like DIA.
130
Alison Chaves @alisonchaves.bsky.social · 10/08/2026
How to ruin an expensive experiment: "Cellular proteins were harvested from three biological replicates at each condition, and the proteins were pooled into one sample for further sample preparation."
020
Alison Chaves @alisonchaves.bsky.social · 08/08/2026
As far as I can tell, I just received the best peer-review criticism of my life. Real improvement suggestions. Things I wasn't aware of when making some decisions. I couldn't be happier. This is a sincere thanks to reviewers.
110
Alison Chaves @alisonchaves.bsky.social · 07/08/2026
Our latest work is out in Protein Science. Check the effects of B. jararaca (snake) envenomation in plasma. onlinelibrary.wiley.com/share/M2WGBJ...
onlinelibrary.wiley.com
Systemic impact of intramuscular snake envenomation and intravenous immunoglobulin‐based antivenom treatment on murine plasma proteome and peptidome
Reader environment loaded
010
Alison Chaves @alisonchaves.bsky.social · 07/08/2026
Oh, very interested in hearing from you. We do acid precipitation followed by SP3 or stagetip. Sometimes, 3kDa filter for plasma peptidomics. All of them increase pressure.
100
Alison Chaves @alisonchaves.bsky.social · 06/08/2026
Expectations in de novo search field: 1. Cascadia working; 2. Casanovo reporting RT and making non-tryptic ckpt compatible with the last release; 3. InstaNovo with winnow integrated; 4. GUI for InstaNovo; 5. Fragnovo release in FragPipe; 6. this is one interesting: langtaosha.org.cn/lts/en/prepr...
langtaosha.org.cn
Accurate and ultra-fast de novo HLA-I immunopeptide sequencing with FoxNovo
We present FoxNovo, a hybrid deep learning-combinatorial framework for de novo sequencing of immunopeptides trained on a large-scale HLA-I immunopeptidomics dataset assembled and reprocessed from publ
021
Alison Chaves @alisonchaves.bsky.social · 05/08/2026
If you are a reviewer, please don't request that the authors cite your work. When the field is niche, we know who you are!
020
Alison Chaves @alisonchaves.bsky.social · 29/07/2026
Evolution can reuse proteins to do other things, as well. Find new utilities to old tools. journals.plos.org/plosbiology/...
journals.plos.org
Evolutionary novelty in gravity sensing through horizontal gene transfer and high-order protein assembly
Author summary A central question in evolutionary biology is how novel traits arise. Gravitropism—the orientation of growth based on gravity—evolved independently in different groups of plants and fun...
020
Alison Chaves @alisonchaves.bsky.social · 27/07/2026
Thank you for the services. 🫡
001
Alison Chaves @alisonchaves.bsky.social · 25/07/2026
This kind of lecture available for free is amazing.
000
Alison Chaves @alisonchaves.bsky.social · 25/07/2026
I don't know if Professor Hamkins is on bsky, but you definitely need to watch his lectures available on YouTube.https://www.youtube.com/watch?v=4Y9p0yp8Mow
youtube.com
Zeno's Paradox and Infinite Sums — Lectures on Infinity (Lecture 1)
YouTube video by Ergo
100
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
That is how we know a clinical trial is underpowered or not.
010
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
Why not? You don't need risk or hazard ratio to calculate it. All you need is Cohen's d, alpha, N, and desired power information.
110
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
I would add: if that subtle difference is to be considered seriously, we should ask for a minimum effect size (MES) calculation. If the difference in protein abundance between groups is greater than the MES, then we care about it.
100
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
To be honest, I don't think the difference is your PCAs are significant. The CI95 are slightly overlapping in one and maybe unrealistically separated on another. Extract the scores from PCA and check the proteins. Check beyond the first 2 PCs. This would suffice for a human cohort.
100
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
Every choice matters. If you use MAD, cyclicloess, VSN or quantile for normalization. If you do imputation (several options). In which order you apply the transformation. If you use PCA, PLS-DA or EFA for clustering.
110
Alison Chaves @alisonchaves.bsky.social · 24/07/2026
If you work in the field of medical mycology, I have a review article to share with you. It’s from 2021, but still worth reading. doi.org/10.1080/1040...
doi.org
What is the elephant in the room when considering new therapies for fungal diseases?
The global scenario of antimicrobial resistance is alarming, and the development of new drugs has not appeared to make substantial progress. The constraints on drug discovery are due to difficultie...
000
Alison Chaves @alisonchaves.bsky.social · 19/07/2026
It would be kind to the trophy.
000
Alison Chaves @alisonchaves.bsky.social · 19/07/2026
Better football 🏆
000
Alison Chaves @alisonchaves.bsky.social · 19/07/2026
Argentina is confounding this with UFC.
100
Alison Chaves @alisonchaves.bsky.social · 07/07/2026
N-Terminal Coumarin Derivatization-Aided De Novo Peptide Sequencing and Its Application to Peptidomics Using LC-Trapped Ion Mobility Spectrometry-qTOF/MS | Analytical Chemistry pubs.acs.org/doi/10.1021/...
pubs.acs.org
N-Terminal Coumarin Derivatization-Aided De Novo Peptide Sequencing and Its Application to Peptidomics Using LC-Trapped Ion Mobility Spectrometry-qTOF/MS
De novo peptide sequencing using LC–MS/MS enables database-independent sequence determination and facilitates the discovery of bioactive peptides and biomarkers in untargeted peptidomics. However, conventional MS/MS analyses often produce fragment ions that are insufficient for complete sequencing. Here, we established an LC–MS/MS workflow using N-succinimidyl 7-methoxycoumarin-3-carboxylate (Me-Cou) as an N-terminal tag to improve de novo sequencing performance. Me-Cou derivatization generated highly informative fragment ions, particularly b1–b3 ions, for (Gly)4 at 1 μmol/L, surpassing the performance of established tags, such as 2,4,6-trinitrobenzenesulfonic acid, 3-aminopyridyl-hydroxy-succinimidyl carbamate, and N-succinimidyl [tris(2,4,6-trimethoxyphenyl)phosphonio]acetate bromide. The Me-Cou tag also enabled confident sequencing of (Gly)10, generating abundant b1–b9 ions. Notably, intact de novo sequencing identified 74 peptides with 32 misidentified peptides that were not included in a 132-standard peptide mixture (86 dipeptides and 46 oligopeptides). However, all peptides were correctly identified using Me-Cou-aided de novo sequencing with no misidentification, demonstrating high specificity and accuracy. Additionally, Me-Cou-aided de novo sequencing successfully identified 328 peptides in casein peptone, all of which were assigned as casein protein fragments. Overall, Me-Cou-based LC–MS/MS significantly enhanced peptide detection and the characterization of sequence diversity compared with intact analysis. This methodology is a promising strategy for untargeted peptidomics, enabling comprehensive characterization of endogenous peptides in both protein hydrolysates and biological fluids.
010
Alison Chaves @alisonchaves.bsky.social · 05/07/2026
Why so many bold text at the introduction? anyway, I'm skeptical that the rejection was based only on the one reviewer's comment saying it wasn't real biology! While the work seems to be legit, there is huge money involved, some commercial interests that need to be disclosed (Biotic company?)
000
Alison Chaves @alisonchaves.bsky.social · 03/07/2026
Definitely, the publication and sharing of code is amazing.
010
Alison Chaves @alisonchaves.bsky.social · 03/07/2026
👏 thanks for the effort in the glycoMS. This is the clever idea I have seen this year.
110
Alison Chaves @alisonchaves.bsky.social · 01/07/2026
If you are looking for an easy way to compute iBAQ from DIA-NN output, ibaq4diann R package is for you. Try it out at: github.com/41ison/ibaq4...
130
Alison Chaves @alisonchaves.bsky.social · 01/07/2026
If you find a smile plot, it is probably Olink.
120
Alison Chaves @alisonchaves.bsky.social · 20/06/2026
"Under identical conditions, signals corresponding to 20 proteinogenic amino acids, 4 post-translationally modified amino acids, 32 peptides, 6 modified peptides, 11 bioactive peptides and 2 neoantigen peptides were recorded." www.nature.com/articles/s41...
nature.com
High-resolution nanopore peptide sensing, profiling and sequence assembly - Nature Nanotechnology
A nickel-modified nanopore enables simultaneous, unambiguous identification of amino acids and peptides. Direct analysis or hydrolysis-based sequencing can be used for profiling and reconstruction, an...
000
Alison Chaves @alisonchaves.bsky.social · 17/06/2026
I hope they don't use Thermo antibodies!
000
Alison Chaves @alisonchaves.bsky.social · 16/06/2026
The following fragment of code if from fragpipe de novo. Does that mean that de novo in fragpipe will run DIA? private static final String[] MODEL_OPTIONS = { "DDA_MassIVE", "ddaPASEF", "DIA_Orbitrap", "diaPASEF", "nonspecific" }
000
Alison Chaves @alisonchaves.bsky.social · 15/06/2026
it is a critique gradient if you reorder 2 3 1 😂
010
Alison Chaves @alisonchaves.bsky.social · 11/06/2026
High pressure at the screen
010
Alison Chaves @alisonchaves.bsky.social · 11/06/2026
I liked this one.
000
Alison Chaves @alisonchaves.bsky.social · 10/06/2026
Well, if that's true, the Waters mass specs remain where they are in proteomics. Nowhere.
010
Alison Chaves @alisonchaves.bsky.social · 10/06/2026
You just created big expectations in all of us.
030
Alison Chaves @alisonchaves.bsky.social · 10/06/2026
Sage is só fast that it sounds like an ilegal thing 😂
040
Alison Chaves @alisonchaves.bsky.social · 09/06/2026
~200 Hz at 100k resolution doesn't seem to be realistic. People say the Xevo could scan at >100 Hz with good results, but nobody can provide a good example of it.
110
Alison Chaves @alisonchaves.bsky.social · 08/06/2026
That was a good one
000