Volume 21 Issue 1, January 2024

Scientists have successes to celebrate but must also cope with the sting of failures. In the way she handles both, Nobel laureate Katalin Karikó inspires others.

It is the mark of an educated mind to rest satisfied with the degree of precision that the nature of the subject admits and not to seek exactness where only an approximation is possible. —Aristotle

Facemap tracks keypoints on the mouse face and feeds the information into a deep neural network to predict neural activity.

As money pours into aging research, the field can combine its many methods to home in on what underpins aging. Approaches differ, but researchers share the desire to not overpromise quick-fix anti-aging methods.

Two studies show that nanopores can identify the 20 proteinogenic amino acids and some of their post-translational modifications. Coupled with an exopeptidase, a bottom-up approach to protein sequencing using nanopores is on the horizon.

How accurate is the prediction of protein structure by AlphaFold? Terwilliger et al. address this question with a rigorous assessment of the accuracy of AlphaFold-predicted structures by comparing them with experimentally determined X-ray crystallographic data.

We developed MAbID, a method for combined genomic profiling of histone modifications and chromatin-binding proteins in single cells, enabling researchers to study the interconnectivity between gene-regulatory mechanisms. We demonstrated MAbID’s implementation in profiling multifactorial changes in chromatin signatures during in vitro neural differentiation and in primary mouse bone marrow tissue.

We developed a machine learning model, RoseTTAFoldNA, that can predict the structures of protein–DNA and protein–RNA complexes. Our model is capable of predicting accurate structures of protein families for which structural information is unknown.

In 1858, the first standard for microscope objectives was established to encourage interchangeable components. Over the following 150 years, standards have evolved to constrain the size of objectives, which limits the parameters of working distance, field of view and resolution. A new design breaks out of this conventional envelope, offering an ultra-long working distance in air and enabling new neuroscience experiments.

Tracking cells is a time-consuming part of biological image analysis, and traditional manual annotation methods are prohibitively laborious for tracking neurons in the deforming and moving Caenorhabditis elegans brain. By leveraging machine learning to develop a ‘targeted augmentation’ method, we substantially reduced the number of labeled images required for tracking.

This paper proposes quantitative modeling of single-cell ATAC-seq data, which improves various downstream analyses.

This study demonstrates the need and advantage of uniformly quantifying single-nucleus ATAC-seq data using Paired-Insertion Counting.

This manuscript describes a refinement protocol that extends the e2gmm method to optimize both the orientation and conformation estimation of particles to improve the alignment for flexible domains of proteins.

By combining fast lift-over and selective re-mapping, levioSAM2 enables efficient and accurate read mapping and variant calling leveraging complete reference genomes.

veloVI enhances RNA velocity analysis with uncertainty quantification and extensibility by deep generative modeling of gene-specific transcriptional dynamics.

An integrative framework to simultaneously interrogate the dynamics of the transcriptome and proteome at subcellular resolution that combines two methods, localization of RNA (LoRNA) and a streamlined density-based localization of proteins by isotope tagging (dLOPIT).

MAbID offers a multiplexing approach to uncover the genomic distributions of various epigenetic markers, enabling the study of how these markers jointly direct gene expression.

ANCOM-BC2 is developed to perform multigroup differential abundance analysis and allows modeling of covariates and longitudinal measures while controlling false discovery rate (FDR) or mixed directional FDR.

A Ni²⁺-modified MspA nanopore construct can unambiguously discriminate the 20 proteogenic amino acids as well as several post-translational modifications.

A phenylalanine-containing peptide probe can be used for discriminating all 20 amino acids via current blockage during translocation through an α-hemolysin (αHL) nanopore. The paper provides proof-of-concept peptide sequencing demonstrations.

An analysis of AlphaFold protein structure predictions shows that while in many cases the predictions are highly accurate, there are also many instances where the predicted structures or parts of predicted structures do not agree with experimentally resolved data. Therefore, care must be taken when using these predictions for informing structural hypotheses.

RoseTTAFoldNA extends the RoseTTAFold2 platform to predict the structures of protein–DNA and protein–RNA complexes.

DeepMainmast is a protein structure modeling protocol for cryo-EM that combines the strengths of a deep-learning-based de novo protein main-chain-tracing approach with AlphaFold2-based structure predictions for improved performance.

The Cousa objective is an ultra-long working distance air objective optimized for two- and three-photon imaging. Bypassing challenges caused by water immersion and short working distances, the Cousa enables and improves imaging of diverse specimens.

Targettrack is a deep-learning-based pipeline for automatic tracking of neurons within freely moving C. elegans. Using targeted augmentation, the pipeline has a reduced need for manually annotated training data.