@lukas80455.bsky.social
Reposted
Solving the computational challenge of phosphoproteomics with 𝐏𝐡𝐨-𝐓𝐢𝐩: dephosphorylation on-tip identifies the sequences of phosphorylated peptides.
This serves as a basis for predicted spectral libraries, reducing the search space 10x-20x. www.biorxiv.org/content/10.1...
This serves as a basis for predicted spectral libraries, reducing the search space 10x-20x. www.biorxiv.org/content/10.1...
November 19, 2025 at 10:07 AM
Solving the computational challenge of phosphoproteomics with 𝐏𝐡𝐨-𝐓𝐢𝐩: dephosphorylation on-tip identifies the sequences of phosphorylated peptides.
This serves as a basis for predicted spectral libraries, reducing the search space 10x-20x. www.biorxiv.org/content/10.1...
This serves as a basis for predicted spectral libraries, reducing the search space 10x-20x. www.biorxiv.org/content/10.1...
Reposted
Fantastic project led by @bo-wen.bsky.social. Excited to see the future uses of AI and transfer learning in proteomics. #massspec #proteomics
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Carafe enables high quality in silico spectral library generation for data-independent acquisition proteomics - Nature Communications
Accurate spectral libraries are essential for analyzing data-independent acquisition (DIA) proteomics data. Here, the authors present Carafe, which trains on DIA data to build experiment-specific spec...
www.nature.com
November 6, 2025 at 3:49 PM
Fantastic project led by @bo-wen.bsky.social. Excited to see the future uses of AI and transfer learning in proteomics. #massspec #proteomics
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted
GO FAST with InfinDIA!! proteomicsnews.blogspot.com/2025/10/tire...
Tired of slow DIA search times? Try DIA-NN 2.3 with InfinDIA!
Is your awesome new instrument generating so much data that you can't process it fast anymore? My data density has jumped about 5x from my...
proteomicsnews.blogspot.com
October 2, 2025 at 11:01 AM
GO FAST with InfinDIA!! proteomicsnews.blogspot.com/2025/10/tire...
Reposted
I will be at the 𝐃𝐈𝐀-𝐍𝐍 session @ 𝐓𝐡𝐞𝐫𝐦𝐨 𝐏𝐫𝐨𝐭𝐞𝐨𝐦𝐞 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐞𝐫 & 𝐂𝐨𝐦𝐩𝐨𝐮𝐧𝐝 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐞𝐫 User Meetings in Bremen, Germany, 𝟗-𝟏𝟏 𝐃𝐞𝐜𝐞𝐦𝐛𝐞𝐫. With some cool advances for DIA-NN on Thermo instruments. www.thermofisher.com/de/de/home/e...
LinkedIn
This link will take you to a page that’s not on LinkedIn
lnkd.in
October 1, 2025 at 7:19 AM
I will be at the 𝐃𝐈𝐀-𝐍𝐍 session @ 𝐓𝐡𝐞𝐫𝐦𝐨 𝐏𝐫𝐨𝐭𝐞𝐨𝐦𝐞 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐞𝐫 & 𝐂𝐨𝐦𝐩𝐨𝐮𝐧𝐝 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐞𝐫 User Meetings in Bremen, Germany, 𝟗-𝟏𝟏 𝐃𝐞𝐜𝐞𝐦𝐛𝐞𝐫. With some cool advances for DIA-NN on Thermo instruments. www.thermofisher.com/de/de/home/e...
Reposted
Precisely mapping proteases specificity and predicting cleavage rates - with peptide libraries: doi.org/10.1101/2025.... This is a powerful application that is uniquely enabled by the peptidoform confidence and QuantUMS modules in DIA-NN 2.0.
October 7, 2025 at 12:33 PM
Precisely mapping proteases specificity and predicting cleavage rates - with peptide libraries: doi.org/10.1101/2025.... This is a powerful application that is uniquely enabled by the peptidoform confidence and QuantUMS modules in DIA-NN 2.0.
Reposted
Wrote a short perspective: Pre-trained DIA models can 'covertly' inflate FDR - and this will not show up in entrapment validation
www.linkedin.com/pulse/pre-tr...
www.linkedin.com/pulse/pre-tr...
Pre-trained DIA models can 'covertly' inflate FDR - and this will not show up in entrapment validation
Transformer models are great? For several years, there has been this idea that one can pre-train advanced deep learning models, e.g.
www.linkedin.com
October 9, 2025 at 11:41 AM
Wrote a short perspective: Pre-trained DIA models can 'covertly' inflate FDR - and this will not show up in entrapment validation
www.linkedin.com/pulse/pre-tr...
www.linkedin.com/pulse/pre-tr...
Reposted
Reposted
With 𝗗𝗜𝗔-𝗡𝗡 𝟮.𝟯.𝟬 Preview (Academia-only for now), we showcase the transformative new capabilities that have been developed in the past months. Download: github.com/vdemichev/Di...
September 26, 2025 at 9:52 AM
With 𝗗𝗜𝗔-𝗡𝗡 𝟮.𝟯.𝟬 Preview (Academia-only for now), we showcase the transformative new capabilities that have been developed in the past months. Download: github.com/vdemichev/Di...
Reposted
Chemoproteomics: can we trust the search engine? Since inception of DIA-NN, we always prioritised data quality. A nice benchmark by @anamdv.bsky.social and colleagues:
www.linkedin.com/posts/vadim-...
www.linkedin.com/posts/vadim-...
𝐖𝐡𝐲 𝐃𝐈𝐀-𝐍𝐍 𝐢𝐬 𝐠𝐫𝐞𝐚𝐭 𝐟𝐨𝐫 𝐜𝐡𝐞𝐦𝐨𝐩𝐫𝐨𝐭𝐞𝐨𝐦𝐢𝐜𝐬?
You can analyse almost anything with it, and with high data quality:
Below is an excellent benchmark by Ana Martínez del Val… | Vadim Demichev
𝐖𝐡𝐲 𝐃𝐈𝐀-𝐍𝐍 𝐢𝐬 𝐠𝐫𝐞𝐚𝐭 𝐟𝐨𝐫 𝐜𝐡𝐞𝐦𝐨𝐩𝐫𝐨𝐭𝐞𝐨𝐦𝐢𝐜𝐬?
You can analyse almost anything with it, and with high data quality:
Below is an excellent benchmark by Ana Martínez del Val and colleagues. Bottom panel (c):...
www.linkedin.com
September 9, 2025 at 7:15 AM
Chemoproteomics: can we trust the search engine? Since inception of DIA-NN, we always prioritised data quality. A nice benchmark by @anamdv.bsky.social and colleagues:
www.linkedin.com/posts/vadim-...
www.linkedin.com/posts/vadim-...
Reposted
Reposted
We have updated our Slice-PASEF preprint! It's a result of a large collaboration where we highlight its advantages for single cell proteomics. Or any other applications benefiting from sensitivity. For example, major gains for ubiquitinomics:
September 2, 2025 at 5:58 PM
We have updated our Slice-PASEF preprint! It's a result of a large collaboration where we highlight its advantages for single cell proteomics. Or any other applications benefiting from sensitivity. For example, major gains for ubiquitinomics:
Reposted
Mathias Wilhelm, @martinfrejno.bsky.social from MSAID as well as Marie Locard-Paulet, @magnuspalmblad.bsky.social
and myself are organising a workshop on FDR control & trustworthiness of proteomics, hosted by Lorentz Center in Leiden, February 2-6, 2026. Stay tuned :)
and myself are organising a workshop on FDR control & trustworthiness of proteomics, hosted by Lorentz Center in Leiden, February 2-6, 2026. Stay tuned :)
Center for Scientific Workshops in All Disciplines - Trustworthiness in Proteomics
www.lorentzcenter.nl
June 4, 2025 at 4:38 PM
Mathias Wilhelm, @martinfrejno.bsky.social from MSAID as well as Marie Locard-Paulet, @magnuspalmblad.bsky.social
and myself are organising a workshop on FDR control & trustworthiness of proteomics, hosted by Lorentz Center in Leiden, February 2-6, 2026. Stay tuned :)
and myself are organising a workshop on FDR control & trustworthiness of proteomics, hosted by Lorentz Center in Leiden, February 2-6, 2026. Stay tuned :)
Reposted
Happy to announce that the development version of DIA-NN has preliminary support for multi-column setups that allow acquiring multiple samples in the same run, in collaboration with @coonlab.bsky.social. See SynchroSep-MS poster by Noah Lancaster for the concept.
June 2, 2025 at 2:17 PM
Happy to announce that the development version of DIA-NN has preliminary support for multi-column setups that allow acquiring multiple samples in the same run, in collaboration with @coonlab.bsky.social. See SynchroSep-MS poster by Noah Lancaster for the concept.
Reposted
DIA-NN 2.2.0 is released! The new DIA-NN 2.2.0 Enterprise achieves up to ~1.6x median speedup on 64-cores under Windows (please see benchmark below). Release notes: github.com/vdemichev/Di..., download (Academia): github.com/vdemichev/Di...
May 30, 2025 at 8:36 AM
DIA-NN 2.2.0 is released! The new DIA-NN 2.2.0 Enterprise achieves up to ~1.6x median speedup on 64-cores under Windows (please see benchmark below). Release notes: github.com/vdemichev/Di..., download (Academia): github.com/vdemichev/Di...
Reposted
DIA-NN 2.0 release is almost ready, will come with some big news. One is scanning methods support, at the moment doing some tuning of the algorithm using Synchro-PASEF data :) Turns out, 'Q1 information' is really helpful for gaining peptidoform confidence. Likely also channel-confidence in plexDIA.
January 9, 2025 at 12:16 PM
DIA-NN 2.0 release is almost ready, will come with some big news. One is scanning methods support, at the moment doing some tuning of the algorithm using Synchro-PASEF data :) Turns out, 'Q1 information' is really helpful for gaining peptidoform confidence. Likely also channel-confidence in plexDIA.
Reposted
You can play with the QuantUMS filters using a dashboard for DIANN
github.com/41ison/QC4DI...
github.com/41ison/QC4DI...
QC4DIANN/QC_report_DIANN_app.r at main · 41ison/QC4DIANN
Quality control report of DIANN 1.9.2 search results - 41ison/QC4DIANN
github.com
December 10, 2024 at 2:18 AM
You can play with the QuantUMS filters using a dashboard for DIANN
github.com/41ison/QC4DI...
github.com/41ison/QC4DI...
Reposted
Sitting at Pike Place Market in Seattle at 7am in the morning, sipping coffee with my son, a long layover on the way home from Taipei. Otherwise I would do a long post. DiaTracer in #FragPipe work really well, making library-free analysis of any diaPASEF data possible. www.nature.com/articles/s41...
diaTracer enables spectrum-centric analysis of diaPASEF proteomics data - Nature Communications
Data-independent acquisition advances proteomics quantification. Here, the authors present diaTracer, a spectrum-centric tool for diaPASEF data that supports broad proteomics applications, enabling di...
www.nature.com
January 4, 2025 at 4:04 PM
Sitting at Pike Place Market in Seattle at 7am in the morning, sipping coffee with my son, a long layover on the way home from Taipei. Otherwise I would do a long post. DiaTracer in #FragPipe work really well, making library-free analysis of any diaPASEF data possible. www.nature.com/articles/s41...