Bioinformatics

Protein Set Transformer: a protein-based genome language model to power high-diversity viromics

1. Cody Martin
2. Anthony Gitter
3. Karthik Anantharaman

Reviewed by Arcadia Science

Autoprot: Processing, Analysis and Visualization of Proteomics Data in Python

1. Julian Bender
2. Wignand W. D. Mühlhäuser
3. Johannes P. Zimmerman
4. Friedel Drepper
5. Bettina Warscheid

Reviewed by Arcadia Science

GenomeDelta: detecting recent transposable element invasions without repeat library

1. Riccardo Pianezza
2. Anna Haider
3. Robert Kofler

Reviewed by Arcadia Science

Limitations of current machine learning models in predicting enzymatic functions for uncharacterized proteins

1. Valérie de Crécy-Lagard
2. Raquel Dias
3. Nick Sexson
4. Iddo Friedberg
5. Yifeng Yuan
6. Manal A Swairjo

Reviewed by Arcadia Science

Exploring the accuracy of ab initio prediction methods for viral pseudoknotted RNA structures

1. Vasco Medeiros
2. Jennifer M. Pearl
3. Mia Carboni
4. Ece Er
5. Stamatia Zafeiri

Reviewed by PREreview

Design of linear and cyclic peptide binders of different lengths from protein sequence information

1. Qiuzhen Li
2. Efstathios Nikolaos Vlachos
3. Patrick Bryant

Reviewed by PREreview

Deep geometric framework to predict antibody–antigen binding affinity

1. Nuwan Bandara
2. Dasun Premathilaka
3. Sachini Chandanayake
4. Sahan Hettiarachchi
5. Vithurshan Varenthirarajah
6. Aravinda Munasinghe
7. Kaushalya Madhawa
8. Subodha Charles

Reviewed by PREreview

FusOn-pLM: A Fusion Oncoprotein-Specific Language Model via Focused Probabilistic Masking

1. Sophia Vincoff
2. Shrey Goel
3. Kseniia Kholina
4. Rishab Pulugurta
5. Pranay Vure
6. Pranam Chatterjee

Reviewed by PREreview

PPIscreenML is a method for structure-based screening of protein-protein interactions using AlphaFold

1. Victoria Mischley
2. Johannes Maier
3. Jesse Chen
4. John Karanicolas

• Curated by eLife

  eLife assessment

  This study explores simple machine learning frameworks to distinguish between interacting and non-interacting protein pairs, offering solid computational results despite some concerns about dataset generation. The authors demonstrate a modest improvement in AlphaFold-multimers' ability to differentiate these pairs. Using a simple yet sound approach, this work is a valuable contribution to the challenging problem of reconstructing protein-protein interaction networks.

Reviewed by eLife

PhysiCell Studio: a graphical tool to make agent-based modeling more accessible

1. Randy Heiland
2. Daniel Bergman
3. Blair Lyons
4. Grant Waldow
5. Julie Cass
6. Heber Lima da Rocha
7. Marco Ruscone
8. Vincent Noël
9. Paul Macklin

• Curated by GigaByte

  Editors Assessment:

  This paper presents a new tool to make using PhysiCell easier, which is an open-source, physics-based multicellular simulation framework with a very wide user base. PhysiCell Studio is a graphical tool that makes it easier to build, run, and visualize PhysiCell models. Over time, it has evolved from being a GUI to include many additional functionalities, and can be used as desktop and cloud versions. This paper outlines the many features and functions, the design and development process behind it, and deployment instructions. Peer review improved the organisation of the various repositories and adding both a requirements.txt and environment.yml files. Looking to the future the developers are planning to add new features based on community feedback and contributions, and this paper presents the many code repositories if readers wish to contribute to the development process.

  This evaluation refers to version 1 of the preprint

Reviewed by GigaByte