Nature - The International Journal of Science / 1 February 2024

7 Research funders must join the fight for equal access to medicines
8 How can scientists make the most of the public’s trust in them?
9 Academia needs radical change — mothers are ready to pave the way
11 Research Highlights
13 Pioneering nuclear-fusion reactor shuts down- what scientists will learn
14 AlphaFold found thousands of possible psychedelics. Will its predictions help drug discovery?
16 Dana-Farber retractions- meet the blogger who spotted problems in dozens of cancer papers
17 Science’s fake-paper problem- high-profile effort will tackle paper mills
18 Japan’s successful Moon landing was the most precise ever
19 Long-COVID signatures identified in huge analysis of blood proteins
20 How does chronic stress harm the gut? New clues emerge
22 How cancer hijacks the nervous system to grow and spread
25 John L. Heilbron (1934–2023), historian of ‘big science’
26 Forget lung, breast or prostate cancer- why tumour naming needs to change
30 Cervical cancer kills 300,000 people a year — here’s how to speed up its elimination
33 Correspondence
35 Ecosystem effects of sea otters limit coastal erosion
36 Mobile atoms enable efficient computation with logical qubits
38 Flexible fibres take fabrics into the information age
39 Bacterial prey turns the tables on predator
40 Snapshots of genetic copy-and-paste machinery in action
42 Contact-tracing app predicts risk of SARS-CoV-2 transmission
45 Designing a circular carbon and plastics economy for a sustainable future
Re-imagining a circular plastics economy
A bold system change is needed
The carbon and plastics life cycle
Key terms of a circular carbon and plastics economy
Evaluating carbon emissions and other metrics
Delivering a bold system change
Sustainable plastics through smart design
Design principles for sustainable plastics
Reduce plastics demand
Switch to renewably sourced plastics
Maximize recycling
Minimize broader environmental impacts
A roadmap towards sustainable plastics
Acknowledgements
Fig. 1 The circular carbon and plastics life cycle.
Fig. 2 Plastic industry scenarios and GHG emissions for 2050 based on an estimated global production of 1.
Fig. 3 The impacts of smart design, service lifespan and recoverability in the carbon footprint of plastics.
Fig. 4 Interventions roadmap for a bold-system-change scenario.
58 Logical quantum processor based on reconfigurable atom arrays
Logical quantum processor based on reconfigurable atom arrays
Logical processor based on atom arrays
Improving entangling gates with code distance
Fault-tolerant logical algorithms
Complex logical circuits using 3D codes
Quantum simulations with logical qubits
Outlook
Online content
Fig. 1 A programmable logical processor based on reconfigurable atom arrays.
Fig. 2 Transversal entangling gates between two surface codes.
Fig. 3 Fault-tolerant logical algorithms.
Fig. 4 Zoned logical processor: scaling and mid-circuit feedforward.
Fig. 5 Complex logical circuits using 3D codes.
Fig. 6 Logical two-copy measurement.
Extended Data Fig. 1 Neutral-atom quantum computer architecture.
Extended Data Fig. 2 Single-qubit Raman addressing.
Extended Data Fig. 3 Mid-circuit readout and feedforward.
Extended Data Fig. 4 Further surface-code data.
Extended Data Fig. 5 Surface-code preparation and decoding data.
Extended Data Fig. 6 [[8,3,2]] and hypercube encoding.
Extended Data Fig. 7 Further [[8,3,2]] circuit sampling data.
Extended Data Fig. 8 Theoretical exploration of hypercube IQP circuits.
Extended Data Fig. 9 Further Bell-basis measurement results.
66 Observation of interband Berry phase in laser-driven crystals
Observation of interband Berry phase in laser-driven crystals
Interband Berry-phase interferometry
Resolving the Berry curvature
Online content
Fig. 1 Interband Berry phase resolved using HHG spectroscopy.
Fig. 2 Berry-phase interferometry.
Fig. 3 Resolving the Berry curvature.
Fig. 4 Circular dichroism HHG spectroscopy.
72High-quality semiconductor fibres via mechanical design
High-quality semiconductor fibres via mechanical design
Stress formation
Capillary instability
Optoelectronic fibres
Applications
Discussion
Online content
Fig. 1 Design and fabrication of semiconductor optoelectronic fibres.
Fig. 2 Stress analysis and capillary instability in the molten-core method.
Fig. 3 Optoelectronic fibres, fabrics and representative applications.
Extended Data Fig. 1 Cracks in the Ge core of Ge/silica fibres.
Extended Data Fig. 2 Comparison of theoretical and FE results on the stress distributions.
Extended Data Fig. 3 Raman spectra for Si/silica and Ge/silica fibres.
Extended Data Fig. 4 Capillary instability in the molten core method.
Extended Data Fig. 5 Glass-clad semiconductor fibres and the removal of glass cladding.
Extended Data Fig. 6 Optoelectronic fibres.
Extended Data Fig. 7 Evolution of the maximum principal stress in the solidified semiconductor core of.
79 Structural transition and migration of incoherent twin boundary in diamond
Structural transition and migration of incoherent twin boundary in diamond
Coexistent multiple ITB configurations
ITB transitions
Configuration-dependent ITB migration
Driving force for ITB activities
Online content
Fig. 1 Coexistent multiple configurations of {112} ITBs in nt-diamond.
Fig. 2 In situ observation of ITB transitions at atomic resolution.
Fig. 3 Configuration-dependent ITB migration.
Fig. 4 Evidence for the stress-driven mechanism of ITB activities.
Extended Data Fig. 1 Microstructural features of nt-diamond.
Extended Data Fig. 2 Structure differences among three symmetrical ITB configurations.
Extended Data Fig. 3 Dislocation characteristics of intrinsic SF, extrinsic SF, and ITB with configuration I.
Extended Data Fig. 4 Dislocations, nanocracks, and boundary disordering associated with ITB transitions.
Extended Data Fig. 5 ITB transition and associated dislocation behaviours observed at the accelerating voltage of 200 kV.
Extended Data Fig. 6 Structural features and relative changes of six ITB configurations, viewed along the (upper) and [111] (lower) zone axes of the left twin domain.
Extended Data Fig. 7 RBD across asymmetric ITBs in diamond.
Extended Data Fig. 8 Mapping of irradiation-induced strains in an nt-diamond grain using PED.
Extended Data Fig. 9 Contour mappings of in-plane atomic strains during an ITB transition from asymmetric configuration V to symmetric configuration III.
Extended Data Fig. 10 Contour mappings of in-plane atomic strains during an ITB transition from mixed IV and VI segments to a hybrid state with multiple configurations f.
86 Durable CO2 conversion in the proton-exchange membrane system
Durable CO2 conversion in the proton-exchange membrane system
Online content
Fig. 1 Physical characterization.
Fig. 2 Electrochemical measurements.
Fig. 3 In situ characterization.
Fig. 4 Theoretical investigation.
92 Stereodivergent 1,3-difunctionalization of alkenes by charge relocation
Stereodivergent 1,3-difunctionalization of alkenes by charge relocation
Online content
Fig. 1 Examples of state-of-the-art remote functionalization reactions of alkenes.
Fig. 2 General reaction scheme, optimization and the scope of syn-selective 1,3-hydroxyacylation.
Fig. 3 Stereodivergence of the 1,3-difunctionalization of alkenes and extension to other product classes.
Fig. 4 Application and mechanistic investigation of the 1,3-difunctionalization of alkenes.
Extended Data Fig. 1 Additional products of syn-selective 1,3-hydroxyacylation.
98 Establishing reaction networks in the 16-electron sulfur reduction reaction
Establishing reaction networks in the 16-electron sulfur reduction reaction
Reaction network in SRR and CV results
In situ Raman study on SRR
The role of catalysis in the SRR network
Conclusion
Online content
Fig. 1 Polysulfide conversion reactions involved in the Li-S battery.
Fig. 2 Charge analysis and reaction network for the SRR.
Fig. 3 In situ Raman results during discharge with the N,S–HGF catalytic electrode.
Fig. 4 Comparison of different catalysts in SRR.
Fig. 5 Simulated site-specific output potential of Li2S4 → Li2S conversion.
105 Flexible silicon solar cells with high power-to-weight ratios
Flexible silicon solar cells with high power-to-weight ratios
Highly efficient flexible and thin SHJ solar cells
Epitaxy-preventing composite gradient passivation
Low-damage continuous-plasma CVD operation
Nanocrystalline sowing and contact vertical growth
Ce-doped indium oxide and laser transfer printing
Cell performance and certification
External quantum efficiency and stability test
Online content
Fig. 1 Schematic diagrams of the FT and SF SHJ solar cells.
Fig. 2 Passivation and nanocrystalline contacts.
Fig. 3 Parameter statistics and certification reports.
Fig. 4 Quantum efficiencies, loss elements and stabilities.
Extended Data Fig. 1 Cross-sectional HRTEM images.
Extended Data Fig. 2 Hydrogen content.
Extended Data Fig. 3 Self-restoring nanocrystalline sowing and vertical growth induction (NSVGI).
Extended Data Fig. 4 Self-restoring nanocrystalline sowing.
Extended Data Fig. 5 Contact-free laser transfer printing.
Extended Data Fig. 6 Endurance assessment.
Extended Data Fig. 7 Durability analysis.
Extended Data Fig. 8 Power recovery in light-induced degradation.
Extended Data Fig. 9 Visualization of the surface potential distribution during RF-PECVD.
Extended Data Table 1 Evaluation of the performance using various technologies.
111Top-predator recovery abates geomorphic decline of a coastal ecosystem
Top-predator recovery abates geomorphic decline of a coastal ecosystem
Long-term relationships
Predator-exclusion experiment
Pre- and post-expansion of sea otters
Spatial comparisons
Discussion
Online content
Fig. 1 Study system and long-term trends in sea otter abundance and creekbank erosion.
Fig. 2 Locations of experimental and observational studies and results from a predator-exclusion experiment.
Fig. 3 Results from an analysis of tidal creeks comparing pre- and post-expansion of sea otters, examining relationships between sea otters, salt marsh biomass and creekbank retreat.
Fig. 4 Relationships between sea otter abundance, shore crab consumption and creekbank erosion in Elkhorn Slough from 2013 to 2015.
Extended Data Fig. 1 Results from a three-year sea otter-exclusion experiment testing the effects of otters and shore crabs on salt marsh vegetation.
Extended Data Fig. 2 Changes in shore crab densities when compared to the first survey in May 2014.
Extended Data Fig. 3 Example of camera-trap data.
Extended Data Fig. 4 Results of a shore crab feeding experiment on pickleweed aboveground and belowground biomass.
Extended Data Fig. 5 Results of a survey examining the relationship between shore crabs and sea otters in tidal creeks, and leverage analysis between sea otter crab consumption and erosion.
Extended Data Fig. 6 Modelling erosion rates.
Extended Data Fig. 7 Illustration of the modelled reduction in the base rate of creek erosion as the number of otters increases.
Extended Data Fig. 8 Posterior parameter distributions from modelling relative creek width as a function of a base rate of widening and an adjustment given the abundance of sea otters.
Extended Data Fig. 9 Histograms of b and r posterior distributions from the second-stage model (while propagating uncertainty from the first model).
Extended Data Fig. 10 Model outputs describing creek changes in width relative to the starting year and erosion accounting for sea otters.
119A lethal mitonuclear incompatibility in complex I of natural hybrids
A lethal mitonuclear incompatibility in complex I of natural hybrids
Mapping mitonuclear incompatibilities
Interactions with X. birchmanni mitochondrial DNA
Lethal effects in early development
Physiology and complex fitness effects
Mitochondrial biology in heterozygotes
Mitonuclear substitutions in complex I
Rapid evolution of complex I proteins
Introgression of incompatibility genes
Discussion
Online content
Fig. 1 Admixture mapping pinpoints mitonuclear incompatibility in Xiphophorus.
Fig. 2 Effect of incompatibility on Xiphophorus hybrid embryos.
Fig. 3 Physiology and proteomics of viable heterozygotes.
Fig. 4 Predicted structures of Xiphophorus respiratory complex I and evolutionary rates of incompatible alleles.
Fig. 5 Phylogenetic analysis and ancestry mapping suggest that genes underlying the mitonuclear incompatibility have introgressed from X.
Extended Data Fig. 1 ABC inference of additional selection parameters.
Extended Data Fig. 2 Ancestry depletion in natural hybrid populations.
Extended Data Fig. 3 Chromosome 15 incompatibility.
Extended Data Fig. 4 Additional F2 embryo morphometrics by ndufs5 genotype.
Extended Data Fig. 5 Juvenile F2 heart morphology by ndufa13 genotype.
Extended Data Fig. 6 Interface between ndufa13, ndufs5, and nd6 in RaptorX model.
Extended Data Fig. 7 Complex I mtDNA gene trees.
Extended Data Table 1 Sensitivity of inter-residue distances to modelling approach.
Extended Data Table 2 CodeML results for Complex I genes.
Extended Data Table 3 SIFT analyses of Complex I genes.
128 Alternative splicing of latrophilin-3 controls synapse formation
Extensive alternative splicing of Lphn3
Lphn3 splicing controls Gαs coupling
Genetic manipulation of Lphn3 splicing
Synapse connection requires Gαs–LPHN3
Assembly of synaptic complexes by LPHN3
Synapse formation requires PBM of E31
Activity promotes E31 splicing
Summary
Online content
Fig. 1 Differentially expressed Lphn3 splice variants couple to different G proteins.
Fig. 2 CRISPR-mediated conversion of Lphn3 alternative splicing from E31 to E32 impairs neuronal network activity.
Fig. 3 Switching LPHN3 G-protein coupling from Gαs to Gα12/13 by deleting E31 suppresses synaptic connectivity of hippocampal neurons.
Fig. 4 The alternatively spliced LPHN3 E31 variant assembles phase-separated postsynaptic scaffold protein condensates.
Fig. 5 Neuronal activity promotes E31 inclusion and E32 exclusion in Lphn3 by alternative splicing, leading to increased expression of the synaptogenic LPHN3 E31 variant.
Extended Data Fig. 1 Alternative splicing of Lphn3 (Adgrl3) transcripts (a & b) and demonstration that a subset of the sites of alternative splicing of Lphn3 exhibits a high degree of cell type-specific expression as revealed by RNAseq analyses (c–e).
Extended Data Fig. 2 Alternative splicing of Lphn1 (Adgrl1), Lphn2 (Adgrl2), and regulation of Lphn1 and Lphn2 G-protein coupling by alternative splicing.
Extended Data Fig. 3 Diverse patterns of Lphn3 alternative splicing analysed by RT-PCR in different brain regions and at different times of postnatal development.
Extended Data Fig. 4 Detailed TRUPATH analyses of G-protein coupling mediated by six different Lphn3 splice variants.
Extended Data Fig. 5 Further data characterizing cAMP assays, RNAseq analyses and pseudorabies virus tracing experiments.
Extended Data Fig. 6 Characterization of purified proteins used for phase-transition experiments.
Extended Data Fig. 7 Independent replication of the Lphn3-E31 dependent recruitment of phase-separated scaffold protein condensates at a higher concentrations of scaffold proteins (38 µM GKAP, 30 µM Homer3, 19 µM PSD95, 32 µM Shank3).
Extended Data Fig. 8 Further characterization of phase-separated, Lphn3-E31 coated post-synaptic scaffold condensates and FRAP (fluorescence recovery after photobleaching) experiments.
Extended Data Fig. 9 Selective deletion of the PDZ-domain binding motif (PBM) in Exon31 of the Lphn3 gene decreases synapse numbers.
Extended Data Fig. 10 Additional analyses of the activity-dependent splicing of Lphn3 Exon31 and Exon32.
136 Cortical regulation of helping behaviour towards others in pain
Cortical regulation of helping behaviour towards others in pain
Helping responses to others in pain
Allolicking helps others cope with pain
Encoding of others’ pain versus stress
Encoding of self-pain versus others’ pain
Control of allolicking and allogrooming
Separable coding of helping and comforting
Discussion
Online content
Fig. 1 Mice exhibit targeted allolicking towards social partners in pain.
Fig. 2 Allolicking by observers reduces self-licking in demonstrators.
Fig. 3 Neural representations of others’ pain and stress in the ACC.
Fig. 4 The ACC bidirectionally regulates allolicking and allogrooming behaviours.
Fig. 5 Separable representation of allolicking and allogrooming in the ACC.
Extended Data Fig. 1 Behavioral responses of demonstrators and observers following melittin injection.
Extended Data Fig. 2 Behaviors of female observers towards female demonstrators in pain.
Extended Data Fig. 3 Observers’ behaviors towards demonstrators experiencing pain induced by formalin injection.
Extended Data Fig. 4 Observers display general allogrooming but not targeted allolicking towards demonstrators in a stress state induced by acute restraint.
Extended Data Fig. 5 Allolicking assists others in coping with pain.
Extended Data Fig. 6 Response of ACC neurons to different states of others across demonstrators.
Extended Data Fig. 7 Single-cell- and population-level representations of prosocial behaviors and different states of demonstrators.
Extended Data Fig. 8 Response of ACC neurons to others’ pain and stress states and during prosocial behaviors.
Extended Data Fig. 9 Behavioral effects of DREADD inhibition of ACC neurons.
Extended Data Fig. 10 Optogenetic activation of ACC neurons and control experiments.
145Digital measurement of SARS-CoV-2 transmission risk from 7 million contacts
Digital measurement of SARS-CoV-2 transmission risk from 7 million contacts
Discussion
Online content
Fig. 1 App risk score and duration of exposure correlate with probability of infection.
Fig. 2 The probability of transmission is affected by both duration and proximity as captured by risk score.
Fig. 3 Transmission probability per exposure window increases almost linearly with risk score.
Fig. 4 Short, intermediate and long exposures all contribute to SARS-CoV-2 transmissions in the population.
Extended Data Fig. 1 The app has more nuanced distance-duration rules than manual contact tracing.
Extended Data Fig. 2 The probability of transmission depends linearly on duration and cumulative risk for short exposures, then sublinearly.
Extended Data Fig. 3 The monotonic relationship between the risk score per window and the probability of transmission in that window is robust with respect to the inclusion of individual heterogeneities in the model.
Extended Data Fig. 4 The transmission probability per exposure window decreases for contacts located in conurbations and increases for low-risk exposures during the weekend.
Extended Data Fig. 5 Duration and cumulative risk are the best predictors of infection, only marginally improved by machine learning.
Extended Data Fig. 6 Illustration of optimal strategies to reduce social costs of contact tracing via amber/red alert notifications.
Extended Data Fig. 7 Illustration of optimal strategies to increase effectiveness of contact tracing via amber/red alert notifications.
Extended Data Table 1 Summary statistics for the NHS COVID-19 app exposure dataset.
Extended Data Table 2 Summary statistics for different types of contacts in our dataset.
151Affinity-optimizing enhancer variants disrupt development
Affinity-optimizing enhancer variants disrupt development
Redundant low-affinity ETS sites regulate the ZRS
Human polydactyly SNVs subtly increase affinity
Affinity-optimizing SNVs cause polydactyly
Predicting penetrance and severity
Affinity-optimizing SNVs prevalent across the ZRS
The enhanceosome contains affinity-optimizing SNVs
Other transcription factors and disease enhancers
Regulatory principles predict causal SNVs
Discussion
Online content
Fig. 1 An ETS-A site in the ZRS enhancer contains two human variants that are associated with polydactyly, both of which subtly increase ETS binding affinity.
Fig. 2 Synthetic changes to the ETS-A site that create a 0.
Fig. 3 All mice with the approximately 0.
Fig. 4 A greater increase in affinity at the ETS-A site causes more severe and penetrant polydactyly and long-bone defects.
Fig. 5 Affinity-optimizing SNVs drive GOF expression in the ZRS and IFNβ enhanceosome.
Fig. 6 Affinity-optimizing SNVs drive GOF expression in a wide variety of disease-associated enhancers.
Extended Data Fig. 1 Class I ETS family members have conserved DBDs.
Extended Data Fig. 2 PBM binding affinities correlate with the in vivo ETS-1 ChIP signal in various cell types.
Extended Data Fig. 3 Conservation of ZRS ETS sites between humans and mice.
Extended Data Fig. 4 EMSA shows the binding of human and mouse ETS-1 to the ETS-A site and ETS-A variants.
Extended Data Fig. 5 Ptch1 in situ hybridization in the hindlimb bud and forelimb bud of transgenic mice.
Extended Data Fig. 6 Syn 0.
Extended Data Fig. 7 Affinity-optimizing SNVs are significantly associated with GOF enhancer activity.
Extended Data Fig. 8 EMSA shows stronger binding of the human HOXA13 and HOXD13 DBDs to the Dutch 2 variant relative to the WT sequence.
Extended Data Fig. 9 Experimental details for MPRA performed with 11 disease-associated enhancers.
Extended Data Fig. 10 Affinity-optimizing eQTL variants are enriched in GOF target gene expression.
160 Autoreactive T cells target peripheral nerves in Guillain–Barré syndrome
Autoreactive T cells in patients with GBS
Cytotoxic TH1 signature of autoreactive T cells
Characterization of autoreactive T cell clones
TCRβ clonotypes in patients with GBS
Antigen recognition and HLA alleles
Autoreactivity in CSF and peripheral nerves
Discussion
Online content
Fig. 1 Ex vivo stimulation of memory CD4+ T cells from the blood of patients with GBS and healthy donors.
Fig. 2 scRNA-seq analysis of memory CD4+ T cells from patients with GBS.
Fig. 3 Characterization of autoreactive CD4+ T cell clones from patients with GBS.
Fig. 4 Clonotypic analysis of autoreactive T cells in patients with GBS.
Fig. 5 Identification of autoreactive CD4+ T cells in the CSF and peripheral nerves of patients with GBS.
Extended Data Fig. 1 Experimental approach for studying autoreactive T cells in patients with GBS.
Extended Data Fig. 2 Overview of the autoreactive memory CD4+ T cell response in patients with GBS, with respect to disease subtype and stage and previous SARS-CoV-2 infection.
Extended Data Fig. 3 Ex vivo stimulation of memory CD8+ T cells from the blood of patients with GBS and healthy donors.
Extended Data Fig. 4 Characterization of autoreactive T cell clones in patients with GBS.
Extended Data Fig. 5 CDR3β consensus motifs characterizing the GLIPH2 specificity clusters.
Extended Data Fig. 6 Association between HLA polymorphisms and public autoreactive TCR Vβ sequences in patients with GBS.
Extended Data Table 1 Patients included in this study.
Extended Data Table 2 List of experiments performed on samples from patients with AIDP.
169 Motion of VAPB molecules reveals ER–mitochondria contact site subdomains
Discussion
Online content
Fig. 1 Altered ER tether motion at ER–mitochondria contact sites.
Fig. 2 Dynamic interactions generate a variable VAPB diffusion landscape within single ER–mitochondria contact sites.
Fig. 3 VAPB contact sites dynamically reorganize according to tether availability and metabolic needs.
Fig. 4 The ALS-linked VAPB P56S mutation displays aberrant motility at ER–mitochondria contact sites.
Extended Data Fig. 1 FIB-SEM reveals association of ER-mitochondrial contact sites and known contact site-associated biology.
Extended Data Fig. 2 Comparisons of VAPB interactions with mitochondria-associated and non-mitochondria associated structures.
Extended Data Fig. 3 Individual ERMCS show variability in capacity to bind passing VAPB molecules.
Extended Data Fig. 4 Latent states can be inferred from trajectory segments with similar mobility profiles.
Extended Data Fig. 5 VAPB-mediated ERMCS contact site mobility is visible in VAPB trajectory analysis.
Extended Data Fig. 6 Stationary contact sites show spatially stable regions of common molecular behaviour.
Extended Data Fig. 7 Changes in ER curvature within ERMCSs.
Extended Data Fig. 8 Variability of VAPB contact site size and shape in well fed and starved cells.
Extended Data Fig. 9 Additional characterization of ERMCS interactions of P56S VAPB molecules.
177 Discovery of a structural class of antibiotics with explainable deep learning
Models for antibiotic activity
Models for human cell cytotoxicity
Filtering and visualizing chemical space
Rationales predict antibiotic classes
Novel filtered substructures
A structural class of antibiotics from rationales
Mechanism of action and resistance
Toxicology, chemical properties and in vivo efficacy
Discussion
Online content
Fig. 1 Ensembles of deep learning models for predicting antibiotic activity and human cell cytotoxicity.
Fig. 2 Filtering and visualizing chemical space.
Fig. 3 Graph-based rationales reveal scaffolds for prospective antibiotic classes.
Fig. 4 Resistance and mechanism of action of a structural class.
Fig. 5 In vivo efficacy.
Extended Data Fig. 1 Molecular weight distribution of the 39,312 compounds screened.
Extended Data Fig. 2 Comparison of deep learning models for predicting antibiotic activity.
Extended Data Fig. 3 Comparison of deep learning models for predicting human cell cytotoxicity.
Extended Data Fig. 4 Visualizing chemical space across different prediction score thresholds.
Extended Data Fig. 5 Examples of rationale calculations using Monte-Carlo tree search.
Extended Data Fig. 6 Maximal common substructure identification reveals known antibiotic classes, but are less predictive than Chemprop rationales across all hits.
Extended Data Fig. 7 Closest active training set compounds to, and selectivities of, four validated hits associated with rationale groups G1-G5.
Extended Data Fig. 8 Comparison of MICs of different compounds against methicillin-susceptible and methicillin-resistant S.
Extended Data Fig. 9 Toxicity, chemical properties, and in vivo efficacy of compounds 1 and 2.
Extended Data Fig. 10 Exploration of a structural class through structure-activity relationships.
186 Template and target-site recognition by human LINE-1 in retrotransposition
Reconstitution of L1 ORF2p-mediated TPRT
Structure of template-RNA-bound L1 ORF2p
Features of the catalytic core
Single-stranded RNA recognition
Novel roles for the C-terminal domain 
Target-site architecture for TPRT
Discussion
Adaptation for nucleic acid recognition
Implications for L1 and SINE lifecycles
Online content
Fig. 1 In vitro TPRT activity and cryo-EM structures of human L1 ORF2p RNPs.
Fig. 2 Recognition of the template RNA and its poly(A) tract.
Fig. 3 Engagement and unwinding of the template RNA by L1 ORF2p CTS.
Fig. 4 Target-site position and upstream single-stranded DNA determine the efficiency of nicking and TPRT.
Extended Data Fig. 1 Purification, electron microscopy and reverse transcriptase activity of human L1 ORF2p and mutants.
Extended Data Fig. 2 Cryo-EM of L1 ORF2p RNP with Alu RNA.
Extended Data Fig. 3 Cryo-EM data processing for L1 ORF2p RNP complex bound to synthetic template RNA.
Extended Data Fig. 4 Resolution estimation.
Extended Data Fig. 5 Active site conformation and supporting data for investigation of the poly(A) tract and stem-loop engagement.
Extended Data Fig. 6 Structure and sequence-based bioinformatics analysis on L1 ORF2p CTS domain.
Extended Data Fig. 7 Analysis of off-target cleavage by L1 ORF2p.
Extended Data Fig. 8 Analysis of target cleavage by ΔCTS L1 ORF2p.
Extended Data Fig. 9 Comparison between the L1 ORF2p RNP and related structures.
Extended Data Fig. 10 Proposed configuration of PCNA interaction with L1 ORF2p.
Extended Data Table 1 Cryo-EM data collection, refinement and validation statistics.
194 Structures, functions and adaptations of the human LINE-1 ORF2 protein
Purification of highly active ORF2p RT
A 2.1 Å crystal structure of the ORF2p core
Five ORF2p core domains all bind nucleic acid
Structure of the L1 wrist domain
ORF2p cryo-electron microscopy structures in three states
Structure of the L1 tower domain
ORF2p RT and polymerase activities
Requirements for ORF2p priming
ORF2p synthesizes cDNA in the cytosol
Synthesized cDNAs activate cGAS/STING
In vitro inhibition of ORF2p
Structural basis of inhibition of ORF2p
Structure of full-length ORF2p
Domain comparison of ORF2p and other RTs
Structural adaptations of ORF2p RT
Structural insight into L1 evolution
Discussion
Online content
Fig. 1 Pathogenic replication cycle of L1 and the 2.
Fig. 2 Cryo-EM structures of ORF2p core in apo, ssRNA and RNA:DNA hybrid-bound states.
Fig. 3 L1 biochemical activities, priming and cytoplasmic reverse transcription of L1.
Fig. 4 Inhibition and structure of full-length ORF2p.
Fig. 5 Structural evolutionary analysis of ORF2p.
Fig. 6 Revised L1 insertion model.
Extended Data Fig. 1 Purification and crystal structure of ORF2p core.
Extended Data Fig. 2 Comparison of cryo-EM maps and models.
Extended Data Fig. 3 Design and characterization of the ORF2p tower domain deletions reveal it is not required for RT.
Extended Data Fig. 4 Priming requirements and mismatch tolerance of ORF2p core.
Extended Data Fig. 5 Comparative enzymology of ORF2p RT with HIV-1 and HERV-K.
Extended Data Fig. 6 Cytoplasmic RT activity of ORF2p and activation of interferon.
Extended Data Fig. 7 Inhibition of ORF2p core by NRTI and NNRTI reverse transcriptase inhibitors.
Extended Data Fig. 8 Comparison of ORF2p with other RTs.
Extended Data Fig. 9 ORF2p and R2Bm structures show opposing topologies of target DNA relative to the active site.
Extended Data Table 1 Data collection and refinement statistics (molecular replacement).
Extended Data Table 2 Cryo-EM Data collection, refinement, and validation statistics.
207 Targeted design of synthetic enhancers for selected tissues in the Drosophila embryo
Online content
Fig. 1 Deep learning-based design of tissue-specific synthetic enhancers.
Fig. 2 Validation of synthetic enhancers in vivo.
Extended Data Fig. 1 Learning the cis-regulatory code of Drosophila embryo tissues with deep learning.
Extended Data Fig. 2 TF motifs predictive of DNA accessibility discovered by TF-Modisco.
Extended Data Fig. 3 Comparison of sequence-to-accessibility and sequence-to-activity models plus controls.
Extended Data Fig. 4 Metric evaluation of the different models.
Extended Data Fig. 5 Predictive value of DNA accessibility and enhancer-activity models for predicted accessible sequences.
Extended Data Fig. 6 Model evaluation on positive and negative control sequences.
Extended Data Fig. 7 Nucleotide contribution scores of synthetic enhancers.
Extended Data Fig. 8 Nucleotide contribution scores of synthetic enhancers.
Extended Data Fig. 9 All synthetic sequences experimentally tested as enhancers.
Extended Data Fig. 10 Predicted scores for synthetic sequences and quantitative validations.
212 Cell-type-directed design of synthetic enhancers
In silico evolutions
Multiple cell-type codes
Motif implantation
Human enhancer design
Discussion
Online content
Fig. 1 Deep learning-based enhancer design.
Fig. 2 In silico sequence evolution towards functional enhancers.
Fig. 3 Spatial expansion and restriction of enhancer activity.
Fig. 4 Motif implantation towards minimal enhancer design.
Fig. 5 Human enhancer design.
Extended Data Fig. 1 In silico sequence evolution from random sequences.
Extended Data Fig. 2 State space optimization, design of perineurial glia enhancers and modification of genomic sequences toward KC enhancers.
Extended Data Fig. 3 Enhancer design toward multiple cell type codes.
Extended Data Fig. 4 Enhancer design by motif implanting.
Extended Data Fig. 5 Human enhancer design by in silico evolution.
Extended Data Fig. 6 Intermediate steps of in silico evolution and generation of repressor sites in human generated enhancers.
Extended Data Fig. 7 Human enhancer design by in silico evolution.
Extended Data Fig. 8 ZEB2 repression of in silico evolved MEL enhancers.
Extended Data Fig. 9 Human enhancer rescue.
Extended Data Fig. 10 Human enhancer design by motif implantation.
221 How co-working labs reduce costs and accelerate progress for biotech start-ups
226 Spreading paws-itivity
e1 Author Correction- A genomic mutational constraint map using variation in 76,156 human genomes
e2 Author Correction- A dense ring of the trans-Neptunian object Quaoar outside its Roche limit
e3 Publisher Correction- Population genomics of post-glacial western Eurasia