Bioinformatics

Convolutional neural networks and outline analyses for archaeobotanical studies of domestication and subspecific identification

1. Vincent Bonhomme
2. Laurent Bouby
3. Julien Claude
4. Camille Dham
5. Muriel Gros-Balthazard
6. Sarah Ivorra
7. Angèle Jeanty
8. Clémence Pagnoux
9. Thierry Pastor
10. Jean-Frédéric Terral
11. Allowen Evin

Reviewed by Peer Community in Archaeology

Subtle methodological variations substantially impact correlation test results in ecological time series

1. Caroline Cannistra
2. Linh Hoang
3. Alex E Yuan
4. Wenying Shou

• Curated by eLife

  eLife Assessment

  This study presents a valuable in-depth comparison of statistical methods for the analysis of ecological time series data, and shows that different analyses can generate different conclusions, emphasizing the importance of carefully choosing methods and of reporting methodological details. The evidence supporting the claims, based on simulated data for a two-species ecosystem, is solid, although testing on more complex datasets could be of further benefit. This paper should be of broad interest to researchers in ecology.

Reviewed by eLife

Decoding molecular mechanisms for loss-of-function variants in the human proteome

1. Matteo Cagiada
2. Nicolas Jonsson
3. Kresten Lindorff-Larsen

• Curated by eLife

  eLife Assessment

  This work introduces FunC-ESMs, a proteome-scale framework to classify loss-of-function missense variants into distinct mechanistic groups by combining two complementary state-of-the-art machine learning models. The strength of evidence is convincing, supported by solid benchmarking, integration with experimental datasets, and careful methodological design. The significance of the findings is valuable, providing a resource of clear interest to researchers and diagnostic laboratories working on variant interpretation.

Reviewed by eLife

Cellpose-SAM: superhuman generalization for cellular segmentation

1. Marius Pachitariu
2. Michael Rariden
3. Carsen Stringer

Reviewed by Arcadia Science

Translational buffering tunes gene expression in mouse and human

1. Shilpa Rao
2. Aden Y Le
3. Logan Persyn
4. Can Cenik

Reviewed by Review Commons

A network regularized linear model to infer spatial expression pattern for single cell

1. Chaohao Gu
2. Hu Chen
3. Zhandong Liu

• Curated by eLife

  eLife Assessment

  The study is useful for advancing spatial transcriptomics through its novel regression-based linear model (glmSMA) that integrates single-cell RNA-seq with spatial reference atlases, and its methodological framework is convincing. The approach demonstrates notable utility by enabling higher-resolution cell mapping across multiple biological systems and spatial platforms compared to existing tools.

Reviewed by eLife

MerQuaCo: a computational tool for quality control in image-based spatial transcriptomics

1. Naomi Martin
2. Paul Olsen
3. Jacob Quon
4. Jazmin Campos
5. Nasmil Valera Cuevas
6. Josh Nagra
7. Marshall VanNess
8. Zoe Maltzer
9. Emily C Gelfand
10. Alana Oyama
11. Amanda Gary
12. Yimin Wang
13. Angela Alaya
14. Augustin Ruiz
15. Cade Reynoldson
16. Cameron Bielstein
17. Christina Alice Pom
18. Cindy Huang
19. Cliff Slaughterbeck
20. Elizabeth Liang
21. Jason Alexander
22. Jeanelle Ariza
23. Jocelin Malone
24. Jose Melchor
25. Kaity Colbert
26. Krissy Brouner
27. Lyudmila Shulga
28. Melissa Reding
29. Patrick Latimer
30. Raymond Sanchez
31. Stuard Barta
32. Tom Egdorf
33. Zachary Madigan
34. Chelsea M Pagan
35. Jennie L Close
36. Brian Long
37. Michael Kunst
38. Ed S Lein
39. Hongkui Zeng
40. Delissa McMillen
41. Jack Waters

• Curated by eLife

  eLife Assessment

  This study provides a valuable contribution to spatial transcriptomics by introducing MerQuaCo, a computational tool for standardizing quality control in image-based spatial transcriptomics datasets. The tool addresses the lack of consensus in the field and provides robust metrics to identify and quantify common imperfections in datasets. The work is supported by an impressive dataset and compelling analyses, and will be of significant interest to researchers focused on data reproducibility and downstream analysis reliability in spatial transcriptomics.

Reviewed by eLife

From Mechanistic Interpretability to Mechanistic Biology: Training, Evaluating, and Interpreting Sparse Autoencoders on Protein Language Models

1. Etowah Adams
2. Liam Bai
3. Minji Lee
4. Yiyang Yu
5. Mohammed AlQuraishi

Reviewed by Arcadia Science

The Implications of Alternative Splicing Regulation for Maximum Lifespan

1. Wei Jiang
2. Sika Zheng
3. Liang Chen

Reviewed by Arcadia Science

Squidly: Enzyme Catalytic Residue Prediction Harnessing a Biology-Informed Contrastive Learning Framework

1. William JF Rieger
2. Mikael Boden
3. Frances Arnold
4. Ariane Mora

• Curated by eLife

  eLife Assessment

  The authors make an important advance in enzyme annotation by fusing biochemical knowledge with language‑model-based learning to predict catalytic residues from sequence alone. Squidly, a new ML method, outperforms existing tools on standard benchmarks and on the CataloDB dataset. The work has solid support, yet clarifications on dataset biases, ablation analyses, and uncertainty filtering would strengthen its efficiency claims.

Reviewed by eLife