In 2006, Shinya Yamanaka and Kazutoshi Takahashi used OCT4, SOX2, KLF4 and MYC (OSKM; also known as Yamanaka factors) to generate induced pluripotent stem cells (iPSCs) from somatic cells. Twenty years later, the field has made remarkable progress, both technically and in terms of its applications. To celebrate the twentieth anniversary of iPSCs, we asked experts around the globe about their view on the most striking achievements and next challenges to overcome.

The basis and physiological functions of senescence have been challenging to establish as senescent states encompass varying hallmarks in disease and ageing, and too little is known about which cells undergo senescence in which contexts. In this Viewpoint, we asked leading researchers in the field to discuss what upcoming questions can help us deepen our knowledge of senescence and what challenges lie ahead.

Recent changes in Genentech’s scientific organization, occurring as the company makes artificial intelligence (AI) increasingly central to its research strategy, highlight a broader question facing biomedical research: how will the growing use of AI reshape the process of scientific discovery?

This Perspective discusses emerging evidence for non-cell-autonomous propagation of ferroptotic cell death and highlights the mechanisms that underlie its restriction in tissues, as well as ferroptosis surveillance beyond the cellular level.

Kulkarni et al. show distinct fates of monomeric versus fibrillar tau in human induced pluripotent stem cell-derived microglia. Monomeric tau is internalized but does not reach degradative lysosomes; fibrillar tau adheres to the cell surface and triggers microglial activation.

Through CRISPR screens and single-molecule imaging, Gupta et al. show that loss of the cohesion acetyltransferase ESCO2 leads to increased G-quadruplex formation at specific regions and drives PARP activation and PARPi hypersensitivity.

Zhang et al. establish a developmentally informed human stem cell model that specifies trophectoderm via Hippo pathway modulation and can self-organize into a three-dimensional model capturing aspects of human chorion development.

Chen, Zhong et al. show that cGAS−STING activation upon chemotherapy treatment leads to TNF expression. Paracrine TNF signalling leads to IKKβ-dependent phosphorylation of ATG9A and autophagy, promoting chemotherapy resistance.

Zheng, Cui, Li et al. demonstrate a role for SMARCAL1 in modulating histone lactylation and chromatin remodelling via phase separation, thereby impacting the antitumour efficacy of radiotherapy and immunotherapy.

The authors show that METTL13-mediated eEF1A K55 dimethylation enhances decoding of rare codons, boosting translation of oncogenic proteins including KRAS and promoting KRAS-inhibitor resistance in lung cancer.

Recent changes in Genentech’s scientific organization, occurring as the company makes artificial intelligence (AI) increasingly central to its research strategy, highlight a broader question facing biomedical research: how will the growing use of AI reshape the process of scientific discovery?