Recep Adiyaman
Weekly Digest March 13, 2026 · 20 min read

Weekly Digest: Mar 09 - Mar 13, 2026

A curated summary of the top protein engineering and structure prediction signals from Mar 09 - Mar 13, 2026.

Share X LinkedIn
Protein Design Daily

Building something in Protein Design?

I love collaborating on new challenges. Let's build together.

Subscribe to Protein Design Digest

Daily curated signals from arXiv, PubMed, and BioRxiv.

🧬 Weekly Recap

Mar 09 - Mar 13, 2026

Missed a day? Here are the top research signals and tools from Monday to Friday, summarized in one place.


🏆 Top Signals of the Week

🗓️ Monday, Mar 09

MutPPI+: a multimodal framework for predicting mutation effects on protein-protein interactions via mutation-path-based data augmentation.

🧬 Abstract

Protein-protein interactions (PPIs) are central to cellular signaling and regulation, and their dysregulation underlies many diseases. Predicting the impact of mutations on PPI stability, quantified as ΔΔG, is essential for understanding disease mechanisms and guiding protein engineering. Here, we first present MutPPI, a graph-based deep-learning model that encodes full-residue structural features of protein-protein complexes and employs a shared GIN-GAT feature extractor for wild-type and mutant complexes. MutPPI outperforms 12 existing methods on an antibody-antigen single-point mutation dataset (S645). By integrating evolutionary information from protein language models, we further develop MutPPI-plus, achieving enhanced predictive performance. Second, we proposed a mutation-path-based data augmentation strategy, which enriches input modalities and improves generalization of both MutPPI and MutPPI-plus. After data augmentation, MutPPI-plus demonstrates state-of-the-art performance on S645 and three additional multi-point mutation datasets (SM_ZEMu, SM595, SM1124), substantially surpassing DDMut-PPI. Our analyses highlight the benefits of the multimodal framework and the physically informed data augmentation method. Together, these results provide a versatile computational tool for accurate ΔΔG prediction, advancing rational protein design.

🗓️ Tuesday, Mar 10

scDock: Streamlining drug discovery targeting cell-cell communication via scRNA-seq analysis and molecular docking.

🧬 Abstract

Summary Identifying drugs that target intercellular communication networks represents a promising therapeutic strategy, yet linking single-cell RNA sequencing (scRNA-seq) analysis to structure-based drug screening remains technically challenging and requires substantial bioinformatics expertise. We present scDock, an integrated and user-friendly pipeline that seamlessly connects scRNA-seq data processing, cell-cell communication inference, and molecular docking-based drug discovery. Through a single configuration file, users can execute the complete workflow, from raw scRNA-seq data to ranked drug candidates, without programming skills. scDock automates the identification of disease-relevant ligand-receptor interactions from scRNA-seq data and performs structure-based virtual screening against these communication targets using Protein Data Bank (PDB) or AlphaFold-predicted protein structures. The pipeline generates comprehensive outputs at each stage, enabling users to explore intercellular signaling alterations and discover therapeutic compounds targeting specific cell-cell communications. scDock addresses a critical gap by providing an accessible end-to-end solution for communication-targeted drug discovery from single-cell data. Availability and implementation scDock is freely available at https://doi.org/10.6084/m9.figshare.31370368 and https://github.com/Andrewneteye4343/scDock. It is implemented in R, Python, shell scripts, and supports Linux systems, including Ubuntu and Debian. Supplementary information Supplementary data are available at Bioinformatics online.

Why it matters: Enhances small-molecule or peptide docking accuracy for targeted drug discovery.

🗓️ Wednesday, Mar 11

Multispectral, Molecular Docking, and Dynamics Simulation Studies of Secalonic Acid F Binding to Human Serum Albumin.

🧬 Abstract

Secalonic acid F (SAF) is a fungal secondary metabolite with broad pharmacological activities. This study investigated the interaction mechanism between SAF and HSA through multispectral techniques, molecular docking, and molecular dynamics simulations. The results show that SAF effectively reduces the intrinsic fluorescence of HSA through static quenching and forms a stable 1:1 molar ratio SAF-HSA complex. SAF binds to the second domain site of HSA. The binding reaction is a spontaneous, exothermic process driven by enthalpy, mainly stabilized through hydrogen bonds and van der Waals forces. Spectral analysis confirmed an increase in the α-helical structure of HSA upon binding. Molecular docking and molecular dynamics simulations, including analyses of RMSD, RMSF, and Rg, further supported and elucidated the experimental results.

🗓️ Thursday, Mar 12

A multimodal approach integrating spectroscopy, deep learning guided molecular docking, and molecular dynamics simulation for predictive assessment of pioglitazone to albumin binding for formulation development.

🧬 Abstract

Binding affinity is a critical parameter that can influence the state of the drug in vivo and help to define the formulation strategy. The current study implements a multimodal approach to analyse the binding affinity between human serum albumin (HSA) and pioglitazone. Ultraviolet (UV) absorbance and fluorescence spectrometry analyses were performed on different combinations of HSA and pioglitazone complexes, and the absorbance and fluorescence intensities were mapped to calculate the binding constant. DynamicBind, a distinct deep-learning artificial intelligence tool, was implemented to perform in silico docking studies using a non-conventional approach. Furthermore, molecular dynamics simulation was also performed to generate root mean square deviation, radius of gyration, and root mean square fluctuation values, followed by principal component analysis, probability distribution function, and free energy landscape analysis. The simulation output was analysed to interpret the binding affinity and associated conformation of the protein-active pharmaceutical ingredient (API) complex. The binding constant calculated through UV analysis was 1.1 × 10 4 M -1 . Fluorescence spectroscopic analysis derived a value of 1.7 × 10 5 M -1 . At the same time, DynamicBind predicted the cLDDT score for the top predicted model to be 0.634, and a binding affinity value of greater than 5, indicating a relatively moderate binding between pioglitazone and HSA. The results from molecular dynamics simulations further complemented our earlier observations, indicating non-covalent binding interactions and a stable protein-API complex, which is desirable for developing a formulation using HSA as a carrier polymer. This orthogonal approach also provided critical information on the fate of the API and possible considerations that needed to be made during the design of the formulation process, highlighting the need for similar approaches that could provide multifaceted advantages and help in optimising R&D costs and timelines.

Why it matters: Expands the searchable sequence space for novel folds and high-affinity binders.

🗓️ Friday, Mar 13

How to make the most of your masked language model for protein engineering

🧬 Abstract

A plethora of protein language models have been released in recent years. Yet comparatively little work has addressed how to best sample from them to optimize desired biological properties. We fill this gap by proposing a flexible, effective sampling method for masked language models (MLMs), and by systematically evaluating models and methods both in silico and in vitro on actual antibody therapeutics campaigns. Firstly, we propose sampling with stochastic beam search, exploiting the fact that MLMs are remarkably efficient at evaluating the pseudo-perplexity of the entire 1-edit neighborhood of a sequence. Reframing generation in terms of entire-sequence evaluation enables flexible guidance with multiple optimization objectives. Secondly, we report results from our extensive in vitro head-to-head evaluation for the antibody engineering setting. This reveals that choice of sampling method is at least as impactful as the model used, motivating future research into this under-explored area.


📚 All Papers & Quick Reads

🗓️ Monday, Mar 09

🗓️ Tuesday, Mar 10

🗓️ Wednesday, Mar 11

🗓️ Thursday, Mar 12

🗓️ Friday, Mar 13


🛠️ Tools & Datasets

  • 🛠 Tool: AlphaFold2 - Deep learning system for high-accuracy protein structure prediction.
  • 🛠 Tool: ColabFold - Fast AlphaFold2/MMseqs2 pipeline for large-scale predictions.
  • 💾 Dataset: UniRef - Clustered protein sequence sets for fast similarity searches.
  • 💾 Dataset: BFD - Big Fantastic Database for deep learning protein modeling.
  • 🛠 Tool: RoseTTAFold - End-to-end neural network for protein structure prediction.
  • 💾 Dataset: MGnify - Metagenomics resource for microbiome sequence data.
  • 🛠 Tool: ESMFold - Language-model-based protein structure prediction from sequences.
  • 💾 Dataset: PDBbind - Binding affinity data with 3D structures of protein-ligand complexes.
  • 🛠 Tool: OmegaFold - Structure prediction from single sequences with rapid inference.
  • 💾 Dataset: BioLiP - Verified biologically relevant ligand-protein interactions.
  • 🛠 Tool: Foldseek - Ultra-fast structural search and clustering engine.
  • 💾 Dataset: SIFTS - Residue-level mapping between PDB, UniProt, and other resources.

🤖 AI in Research Recap


🏢 Industry & Real-World Applications


💼 Jobs & Opportunities


📅 Events


Enjoyed this digest? Subscribe above to get these dailies in your inbox every morning.

BS HF DK