How these papers were chosen among the week's 258 papers

All 258 papers of the week whose main category is in this field were read, title and start of the abstract, and 24 were kept. They were then ordered by what can be counted: how many research profiles of our readers they come close to, whether a rising topic of the field includes them, and how many categories they were posted in. The abstracts of the first 25 were read in full to choose these, no more than two from one category and one per team.

The 3 numbers of the text were checked against the abstracts. 28 statements on what the papers found were looked for word for word in their abstracts.

Sources: bioRxiv, medRxiv.

Sleep opens a window for cellular housekeeping #

Sleep’s restorative effects may reach into neurons’ waste-disposal machinery. In mice, extended wakefulness suppressed autophagic flux—the passage of cellular material through a recycling and disposal system—in cortical neurons. Sleep rapidly restored it, while the material being processed shifted from mitochondrial components towards synaptic proteins. Sleep also increased expression of ATG7, a protein involved in autophagy. Experimentally increasing neuronal activity impaired the process and promoted intracellular amyloid-beta accumulation.

The striking result concerned timing: optogenetically stimulating lysosomal acidification during sleep reduced amyloid pathology in mice modeling Alzheimer’s disease, whereas stimulation during wakefulness worsened it. This preprint suggests that an intervention’s effects can depend on the brain state in which it is delivered, not simply on whether it activates a seemingly beneficial cellular process. The proposed involvement of a noncanonical secretory route remains tentative.

The same brain rhythm, different cognitive effects #

A human stimulation study offers another example of why timing matters. Researchers used EEG to deliver repetitive transcranial magnetic stimulation to the left dorsolateral prefrontal cortex at either the peak or trough of participants’ own beta oscillations. Peak-targeted stimulation improved response inhibition; trough-targeted stimulation improved working memory. Conventional stimulation delivered without reference to the rhythm’s phase served as an active comparison.

The behavioural changes were accompanied by changes in electrical responses and different patterns of network connectivity: peak stimulation strengthened posterior centro-parietal connections, while trough stimulation expanded anterior prefrontal networks. The preprint supports the idea that different phases of an ongoing rhythm help organize different computations. For brain stimulation, it suggests that matching the frequency may not be enough: the point within each cycle could also matter. The abstract does not report participant numbers, leaving the scale of the evidence unclear.

Helping surviving motor pathways carry the signal #

After damage to the corticospinal tract, could stimulation make better use of the connections that remain? Researchers tested this approach in three monkeys with corticospinal tract lesions, stimulating projections from the motor thalamus to primary motor cortex. The animals became better able to perform reaching and fine dexterity tasks, with improvements in strength, gross hand function and digit control.

Electrical recordings suggested that stimulation worked by facilitating surviving descending signals, with stronger effects when the targeted fibres had monosynaptic connections to corticospinal motoneurons. Rather than rebuilding a damaged pathway, the strategy aims to increase the effectiveness of residual circuitry. The use of clinically approved stimulation parameters makes translation a plausible next question, but this preprint does not establish effectiveness in people after stroke.

A genetic distinction in inherited Alzheimer’s disease #

Disease-associated variants in PSEN1 are often discussed in terms of whether they increase or diminish the protein’s function. A new preprint asks a more specific question: does losing a functional copy itself cause early-onset Alzheimer’s disease? Across the Alzheimer’s Disease Sequencing Project, UK Biobank and BioVU, researchers identified 45 carriers of variants predicted to trigger destruction of the gene’s RNA message. None had a recorded case of early-onset dementia; 17 were older than 65 years at last follow-up without reported dementia or Alzheimer’s disease.

The team also re-examined a previously reported stop-gain variant associated with early-onset disease. Their analysis identified a cryptic splice site predicted to produce an in-frame deletion of 5 amino acids. Together, the findings favour a distinction between producing an altered protein and simply failing to produce one. That matters for interpreting genetic tests and disease mechanisms, although predicted molecular effects and the absence of diagnoses in records are not definitive demonstrations of harmlessness.

An evolved attraction that still has to be learned #

House mice produce trimethylamine and approach its smell, unlike the aversive response retained by other mammals, including Mus caroli. Yet this preprint finds that the attraction depends on exposure during development rather than appearing automatically. Whole-brain activity mapping revealed experience-dependent responses, and activity-dependent labelling identified neurons in the anterior olfactory nucleus whose activity was necessary and sufficient for attraction—consistent with a stored memory of the odour.

In Mus caroli, innate avoidance instead depended on excitatory neurons in the posterior cortical amygdala. The comparison challenges a simple equation between evolved behaviour and hard-wired preference: evolution can also produce circumstances in which learning reliably builds a species-typical response. The attraction-related neurons are consistent with a memory engram, though that interpretation is not established conclusively by the abstract.

When mitochondrial trouble activates the wrong defence #

Mitochondrial failure can provoke antiviral responses even without a viral infection. In mice and human brain organoids modeling Leigh syndrome, researchers found an antiviral programme in vulnerable inhibitory neurons. But removing the interferon receptor Ifnar1 or the immune-signalling mediator Sting1 did not improve disease progression. Another sensor, protein kinase R, or PKR, proved more consequential: it associated with mitochondrial RNA, and deleting or silencing it delayed neurological decline, reduced neuroinflammation and extended survival.

Those benefits occurred without restoring mitochondrial respiration. PKR deletion also normalized JNK signalling without detectable global repression of protein translation, suggesting a route beyond PKR’s familiar role in shutting down protein production. The preprint separates an immune response that accompanies disease from a component that helps drive it. It points towards modifying the damaging response to an energy defect, while leaving the underlying respiratory failure unresolved.


Across these preprints, context repeatedly changes the explanation. A cellular intervention can help during sleep and harm during wakefulness; stimulation can favour different cognitive processes at different phases of a rhythm; an evolved preference can depend on experience. The disease studies make a related distinction between an apparent culprit and the mechanism actually doing damage. Together, they argue for asking not only what a molecule or circuit does, but when, through which route, and under what conditions.

Rising in this field

All the rising topics →

Was this letter useful to you?

Written by AI from the 282 papers of the week in this field. Most are preprints, not yet peer reviewed: the text says what their authors report.

Get one letter by email every week with a free account, all of them with Premium. Create a free account