Deep learning is widely applicable to phenotyping embryonic development and disease.

Genome editing simplifies the generation of new animal models for congenital disorders. However, the detailed and unbiased phenotypic assessment of altered embryonic development remains a challenge. Here, we explore how deep learning (U-Net) can automate segmentation tasks in various imaging modalities, and we quantify phenotypes of altered renal, neural and craniofacial development in Xenopus embryos in comparison with normal variability. We demonstrate the utility of this approach in embryos with polycystic kidneys (pkd1 and pkd2) and craniofacial dysmorphia (six1). We highlight how in toto light-sheet microscopy facilitates accurate reconstruction of brain and craniofacial structures within X. tropicalis embryos upon dyrk1a and six1 loss of function or treatment with retinoic acid inhibitors. These tools increase the sensitivity and throughput of evaluating developmental malformations caused by chemical or genetic disruption. Furthermore, we provide a library of pre-trained networks and detailed instructions for applying deep learning to the reader's own datasets. We demonstrate the versatility, precision and scalability of deep neural network phenotyping on embryonic disease models. By combining light-sheet microscopy and deep learning, we provide a framework for higher-throughput characterization of embryonic model organisms. This article has an associated 'The people behind the papers' interview.

Replication Fork Slowing and Reversal upon DNA Damage Require PCNA Polyubiquitination and ZRANB3 DNA Translocase Activity.

DNA damage tolerance during eukaryotic replication is orchestrated by PCNA ubiquitination. While monoubiquitination activates mutagenic translesion synthesis, polyubiquitination activates an error-free pathway, elusive in mammals, enabling damage bypass by template switching. Fork reversal is driven in vitro by multiple enzymes, including the DNA translocase ZRANB3, shown to bind polyubiquitinated PCNA. However, whether this interaction promotes fork remodeling and template switching in vivo was unknown. Here we show that damage-induced fork reversal in mammalian cells requires PCNA ubiquitination, UBC13, and K63-linked polyubiquitin chains, previously involved in error-free damage tolerance. Fork reversal in vivo also requires ZRANB3 translocase activity and its interaction with polyubiquitinated PCNA, pinpointing ZRANB3 as a key effector of error-free DNA damage tolerance. Mutations affecting fork reversal also induced unrestrained fork progression and chromosomal breakage, suggesting fork remodeling as a global fork slowing and protection mechanism. Targeting these fork protection systems represents a promising strategy to potentiate cancer chemotherapy.

DNA2 drives processing and restart of reversed replication forks in human cells.

Accurate processing of stalled or damaged DNA replication forks is paramount to genomic integrity and recent work points to replication fork reversal and restart as a central mechanism to ensuring high-fidelity DNA replication. Here, we identify a novel DNA2- and WRN-dependent mechanism of reversed replication fork processing and restart after prolonged genotoxic stress. The human DNA2 nuclease and WRN ATPase activities functionally interact to degrade reversed replication forks with a 5'-to-3' polarity and promote replication restart, thus preventing aberrant processing of unresolved replication intermediates. Unexpectedly, EXO1, MRE11, and CtIP are not involved in the same mechanism of reversed fork processing, whereas human RECQ1 limits DNA2 activity by preventing extensive nascent strand degradation. RAD51 depletion antagonizes this mechanism, presumably by preventing reversed fork formation. These studies define a new mechanism for maintaining genome integrity tightly controlled by specific nucleolytic activities and central homologous recombination factors.

Human RECQ1 promotes restart of replication forks reversed by DNA topoisomerase I inhibition.

Topoisomerase I (TOP1) inhibitors are an important class of anticancer drugs. The cytotoxicity of TOP1 inhibitors can be modulated by replication fork reversal through a process that requires poly(ADP-ribose) polymerase (PARP) activity. Whether regressed forks can efficiently restart and what factors are required to restart fork progression after fork reversal are still unknown. We have combined biochemical and EM approaches with single-molecule DNA fiber analysis to identify a key role for human RECQ1 helicase in replication fork restart after TOP1 inhibition that is not shared by other human RecQ proteins. We show that the poly(ADP-ribosyl)ation activity of PARP1 stabilizes forks in the regressed state by limiting their restart by RECQ1. These studies provide new mechanistic insights into the roles of RECQ1 and PARP in DNA replication and offer molecular perspectives to potentiate chemotherapeutic regimens based on TOP1 inhibition.