Search

A new suite of studies from the Pan Cancer Analysis of Whole Genomes (PCAWG) Consortium provides the most detailed resolution of cancer genomes to date, extending our knowledge of driver genes, mutational features, structural alterations and more. Kreisberg, Ideker, Mills and Meric-Bernstam discuss the foundational and translational insights gained from this project.

Integrating physical and genetic interaction data is essential if we are to fully understand cellular networks. The classification of interactions beyond the simple physical versus genetic divide promises to accelerate progress, as illustrated by recent successes in network integration.

In cutaneous melanoma, desmosome genes are frequently affected by coding mutations that decrease their expression. This decrease in expression occurs predominantly in keratinocytes in the tumor microenvironment and can induce hyperproliferation of adjacent melanoma cells in coculture.

Trey Ideker and colleagues report a comprehensive genome-wide analysis of head and neck squamous cell carcinoma, reporting that TP53 mutations are frequently accompanied by loss of chromosome 3p. Their data indicate that the combination of these two events has a stronger negative effect on survival rate than either event alone.

Ideker and colleagues present NeST-VVN, a deep learning model based on cancer protein assembly data that can be used to predict the response and resistance of cancer cells to CDK4/6 inhibitors.

Analysis of whole-genome sequences and transcription data from tumors identifies noncoding loci in which mutations affect target gene expression. These somatic eQTLs can classify tumors into pathway-based subtypes and are disrupted in 88% of tumors.

Generative artificial intelligence (AI) can be used to guide cancer drug discovery. Here the authors present a generative AI model that can design drug-like compounds based on specific cancer genotypes.

In tumours aberrant epigenetic modifications can alter the transcriptional state. Here, the authors identify a common tumour-specific shift to transcriptional repression associated with DNA methylation and chromatin dysregulation at the transcription start site.

Network propagation is based on the principle that genes underlying similar phenotypes are more likely to interact with each other. It is proving to be a powerful approach for extracting biological information from molecular networks that is relevant to human disease.

A global map of human subcellular architecture yields protein complex structures, reveals protein functions, identifies assemblies with multiple localizations or cell-type specificity and decodes paediatric cancer genomes.

Network biology is beginning to tackle the complexities of multicellular systems and disease associations.

Large language models show potential in suggesting common functions for a gene set.

A library of plasmids expressing two gRNAs allows for the mapping of combinatorial genetic interactions with the CRISPR system. Results in cancer cells suggest that cellular context is an important factor for the interaction network.

GenomeSpace is an open-source, cloud-based interoperability platform that facilitates integrative genomic analyses, allowing users to transition seamlessly between a diverse and growing set of bioinformatics tools and data resources.

Primary cilia are tiny hair-like structures expressed on the surface of eukaryotic cells. They participate in a range of processes, such as sensing the extracellular environment and regulating signalling pathways during development. Here, a functional genomic screen is presented that used RNA interference to identify human genes involved in controlling ciliogenesis. Several positive and negative ciliogenesis modulators with broad-ranging functions were found.

A framework combining combinatorial CRISPR knockout in breast, lung and oropharyngeal cancer in vitro models with public data identifies synthetic lethal interactions, such as perturbed KDM5C sensitizing cells to PARP7 inhibition.

The recent proliferation of omics data has required a toolbox of integrative systems biology bioinformatics approaches to elucidate functional relationships between molecules. Here the authors explain the principles behind these approaches and discuss their applications.

Polypharmacology drugs are compounds designed to inhibit multiple protein targets. Here, authors use recent advances in AI to rapidly generate polypharmacology compounds against any pair of protein targets, experimentally validating numerous compounds targeting MEK1 and mTOR.

Protein immunofluorescence imaging and affinity purification–mass spectrometry are combined to create a unified map of human cell architecture across scales, which the authors call the multi-scale integrated cell (MuSIC).

This paper reports a strategy for combining somatic mutation profiles of human tumors with gene networks to stratify tumors into biologically and clinically relevant subtypes. The method is applied to ovarian, uterine and lung cancers.

Koch and colleagues report that epigenetic clocks mirror age predictions based on the accumulation of somatic mutations and show that somatic mutations at CpG sites coincide with extensive remodeling of the surrounding methylome.

A human–SARS-CoV-2 protein interaction map highlights cellular processes that are hijacked by the virus and that can be targeted by existing drugs, including inhibitors of mRNA translation and predicted regulators of the sigma receptors.

Next generation sequencing has been pivotal in identifying somatic mutations in cancer samples. In this study, the authors analyse whole exome sequencing data using a variety of bioinformatic methods to identify mutations in known cancer genes and find that many cancer genes are not detected by these methods.

Cell lineage tracing is scaled up to hundreds of millions of simulated sequences with distributed computing.

This paper identified >500 genetic loci associated with behaviors and disorders related to self-regulation, including addiction and child behavior problems. The resulting genetic risk scores predict several behavioral, medical and social outcomes.

There is increasing evidence that protein-coding genes can emerge de novo from noncoding genomic regions. Vakirlis et al. propose that sequences encoding transmembrane polypeptides can emerge de novo in thymine-rich genomic regions and provide organisms with fitness benefits.

Fanconi anemia is a complex disease affecting multiple DNA repair proteins that resolve DNA crosslinks which can block vital processes. Here the authors use parallel genome-wide screens that identify the BLM helicase complex as a suppressor of Fanconi anemia phenotypes.

The CRISPR–Cas9 system enables global screens of gene function with high sensitivity and specificity, but off-target effects have been reported for CRISPR guide RNAs targeting genes that are amplified at high copy number. A new study describes a computational approach to correct for this copy number effect, increasing the specificity of CRIPSR screens to identify essential genes.

Embedding a deep-learning model in the known structure of cellular systems yields DCell, a ‘visible’ neural network that can be used to mechanistically interpret genotype–phenotype relationships.

Informative pathways driving cancer pathogenesis and subtypes can be difficult to identify in the presence of many gene interactions irrelevant to cancer. Here, the authors describe an approach for cancer gene pathway analysis based on key molecular interactions that drive cancer in relevant tissue types, and they assemble a focused map of Evolutionarily Selected Pathways (ESP) with interactions supported by both protein–protein binding and genetic epistasis during somatic tumor evolution.

Different experimental designs for protein interaction mapping are modeled to compare their efficiency in completing an interactome map. Testing of the strategy that minimized the final experimental cost in an ongoing Drosophila melanogaster interactome project found 450 high-confidence interactions using only 47 microtiter plates.

Ma et al. apply few-shot learning to train a neural network model on cell-line drug-response data, and they subsequently transfer it to distinct biological contexts including different tissues and patient-derived tumor cells and xenografts.

Multi-omics analysis of influenza A virus cellular restriction in human macrophage-like cells reveals a mechanism for viral evasion of autophagy.

The small coding capacity of the influenza A virus demands that the virus use the host cellular machinery for many aspects of its life cycle. An integrated systems approach, based on genome-wide RNA interference screening, is now used to identify 295 cellular cofactors required for early-stage influenza virus replication. Knowledge of these host cell requirements provides further targets that could be pursued for antiviral drug development.

Lineage specification and development require a hierarchy of transcription factors. Murre and colleagues have compiled a genome-wide set of cis-acting targets centered on E2A, EBF1 and Foxo1 that govern early B cell development.

This resource describes a human MAP kinase interactome. Yeast two hybrid-derived protein interaction maps of MAP kinase pathway members are refined using multiple criteria, tested against reference sets, and functionally validated using knockdown with siRNA.