Diego Javier Zea
@diegozea.bsky.social
Associate Professor at Université Paris-Saclay. Interested in Protein Structure, Interactions and Evolution, Machine Learning, and
Julia programming.
https://diegozea.github.io/
Julia programming.
https://diegozea.github.io/
October 17, 2025 at 4:30 PM
🛡️ Confidence boost: refreshed benchmarks and tests now cover 96%+ of the code. Repo: github.com/diegozea/MIT...
GitHub - diegozea/MIToS.jl: A Julia package to analyze protein sequences, structures, and evolutionary information
A Julia package to analyze protein sequences, structures, and evolutionary information - diegozea/MIToS.jl
github.com
October 17, 2025 at 4:29 PM
🛡️ Confidence boost: refreshed benchmarks and tests now cover 96%+ of the code. Repo: github.com/diegozea/MIT...
🎨 Better structure viz: edit B-factors so you can color atoms/residues by any variable when plotting (e.g., quality, flexibility, or scores).
October 17, 2025 at 4:29 PM
🎨 Better structure viz: edit B-factors so you can color atoms/residues by any variable when plotting (e.g., quality, flexibility, or scores).
🗂️ Combining datasets is simpler: merge MSAs by matching sequence names automatically when pairings aren’t specified.
October 17, 2025 at 4:29 PM
🗂️ Combining datasets is simpler: merge MSAs by matching sequence names automatically when pairings aren’t specified.
📈 MSA stats at a glance: estimate the effective number of sequences (down-weight redundancy) and quantify the share of “positive” residue pairs under your chosen matrix.
October 17, 2025 at 4:29 PM
📈 MSA stats at a glance: estimate the effective number of sequences (down-weight redundancy) and quantify the share of “positive” residue pairs under your chosen matrix.
📊 Alignment quality: get the classic sum-of-pairs score for an MSA using a residue substitution matrix (BLOSUM62 by default), with consistent handling of gaps and unknowns.
October 17, 2025 at 4:29 PM
📊 Alignment quality: get the classic sum-of-pairs score for an MSA using a residue substitution matrix (BLOSUM62 by default), with consistent handling of gaps and unknowns.
🧬 Substitution matrices are now first-class: use built-ins like BLOSUM62 or bring your own from BioJulia’s BioAlignments.
October 17, 2025 at 4:29 PM
🧬 Substitution matrices are now first-class: use built-ins like BLOSUM62 or bring your own from BioJulia’s BioAlignments.
Join the MAGI team at @i2bcparissaclay.bsky.social as part of the ANR SPPICES project 🌶️
This position involves collaboration with members of the ML4NGP network on machine learning & non-globular proteins
Learn more 👉 sppices.notion.site
This position involves collaboration with members of the ML4NGP network on machine learning & non-globular proteins
Learn more 👉 sppices.notion.site
ANR SPPICES
Scoring and Predicting Protein Interactions and Conformations based on Evolutionary Signals
sppices.notion.site
June 13, 2025 at 11:12 AM
Join the MAGI team at @i2bcparissaclay.bsky.social as part of the ANR SPPICES project 🌶️
This position involves collaboration with members of the ML4NGP network on machine learning & non-globular proteins
Learn more 👉 sppices.notion.site
This position involves collaboration with members of the ML4NGP network on machine learning & non-globular proteins
Learn more 👉 sppices.notion.site
Learn more about the SPPICES project (Scoring and Predicting Protein Interactions and Conformations based on Evolutionary Signals) here:
sppices.notion.site
sppices.notion.site
ANR SPPICES
Scoring and Predicting Protein Interactions and Conformations based on Evolutionary Signals
sppices.notion.site
June 12, 2025 at 9:33 AM
Learn more about the SPPICES project (Scoring and Predicting Protein Interactions and Conformations based on Evolutionary Signals) here:
sppices.notion.site
sppices.notion.site
I have the same question. I haven't checked yet.
March 6, 2025 at 6:16 PM
I have the same question. I haven't checked yet.
Yes, before, it was pretty cumbersome and slow. Now, it is the opposite experience 🙂
March 6, 2025 at 5:24 PM
Yes, before, it was pretty cumbersome and slow. Now, it is the opposite experience 🙂
Since you knew it, could it have helped more unconsciously, being there feeding the intuitions? 🤔
December 7, 2024 at 11:45 AM
Since you knew it, could it have helped more unconsciously, being there feeding the intuitions? 🤔
I am not sure that the Julia's Fibonacci implementation is optimal. Using Val is nice to showcase dispatch, but it hits compilation for every number, and it makes the code type unstable. Conversely, using Val should make the compiler memoize the result for each number. github.com/bddicken/lan...
languages/fibonacci/julia/code.jl at bf71f98f0a823775c16fdb79f2445438f471edf6 · bddicken/languages
Compare languages. Contribute to bddicken/languages development by creating an account on GitHub.
github.com
December 6, 2024 at 7:08 AM
I am not sure that the Julia's Fibonacci implementation is optimal. Using Val is nice to showcase dispatch, but it hits compilation for every number, and it makes the code type unstable. Conversely, using Val should make the compiler memoize the result for each number. github.com/bddicken/lan...