pith. sign in

arxiv: 2511.02824 · v2 · pith:4EIUKAODnew · submitted 2025-11-04 · 💻 cs.AI

Kosmos: An AI Scientist for Autonomous Discovery

Ludovico Mitchener , Angela Yiu , Benjamin Chang , Mathieu Bourdenx , Tyler Nadolski , Arvis Sulovari , Eric C. Landsness , Daniel L. Barabasi
show 29 more authors
Siddharth Narayanan Nicky Evans Shriya Reddy Martha Foiani Aizad Kamal Leah P. Shriver Fang Cao Asmamaw T. Wassie Jon M. Laurent Edwin Melville-Green Mayk Caldas Albert Bou Kaleigh F. Roberts Sladjana Zagorac Timothy C. Orr Miranda E. Orr Kevin J. Zwezdaryk Ali E. Ghareeb Laurie McCoy Bruna Gomes Euan A. Ashley Karen E. Duff Tonio Buonassisi Tom Rainforth Randall J. Bateman Michael Skarlinski Samuel G. Rodriques Michaela M. Hinks Andrew D. White
This is my paper

Pith reviewed 2026-05-22 08:36 UTC · model grok-4.3

classification 💻 cs.AI
keywords AI scientistautonomous discoverystructured world modelhypothesis generationscientific reportsdata analysisliterature searchtraceable reasoning
0
0 comments X

The pith

Kosmos uses a structured world model to keep an AI scientist coherent across 200 rollouts of data analysis and literature search, producing traceable reports that match six months of human research on average.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Kosmos is an AI system that takes an open-ended objective and a dataset then runs repeated cycles of parallel data analysis, literature search, and hypothesis generation before turning the results into scientific reports. The system maintains focus on the original goal over hundreds of individual agent actions by using a shared structured world model that passes information between specialized agents. Each report statement is backed by either executed code or primary literature sources for traceability. Independent scientists judged 79.4 percent of the statements accurate, and human collaborators found that one 20-cycle run produced the equivalent of six months of their own research time while the number of useful findings grew linearly with added cycles. The paper demonstrates this approach through seven discoveries across metabolomics, materials science, neuroscience, and statistical genetics, including three that reproduced results from work the system had not seen.

Core claim

Kosmos automates data-driven scientific discovery by running up to twelve hours of iterative cycles that combine data analysis, literature search, and hypothesis generation. A structured world model shares information between a data analysis agent and a literature search agent so the system can pursue the stated objective coherently across an average of two hundred agent rollouts, collectively executing forty-two thousand lines of code and reading fifteen hundred papers. All statements in the final reports are cited to either code or primary literature. Independent scientists rated 79.4 percent of the statements accurate, and collaborators reported that a single twenty-cycle run performed on

What carries the argument

The structured world model that shares information between the data analysis agent and the literature search agent to preserve coherent pursuit of the objective over many rollouts.

If this is right

  • Valuable scientific findings increase linearly as more cycles are executed up to at least twenty cycles.
  • Reports include citations to code or primary literature for every statement, making the reasoning traceable.
  • Three of the seven highlighted discoveries reproduced results from preprinted or unpublished manuscripts the system had not seen.
  • Four of the discoveries constitute novel contributions to the scientific literature in metabolomics, materials science, neuroscience, and statistical genetics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The linear scaling of findings with cycles implies that longer runs or greater parallel compute could produce proportionally more discoveries without changing the underlying method.
  • The same world-model approach could be tested on objectives that require closed-loop interaction with physical experiments rather than only existing datasets.
  • Because every claim is tied to either code or literature, the reports could serve as starting points for human researchers to verify or extend the findings more quickly.

Load-bearing premise

The structured world model and agent hand-offs can keep the system focused on the original objective across hundreds of steps without introducing systematic biases or hallucinations that the citation system fails to catch.

What would settle it

Run Kosmos on a dataset whose correct discoveries are already published and check whether the generated report independently recovers those same findings while citing only the sources it actually accessed.

read the original abstract

Data-driven scientific discovery requires iterative cycles of literature search, hypothesis generation, and data analysis. Substantial progress has been made towards AI agents that can automate scientific research, but all such agents remain limited in the number of actions they can take before losing coherence, thus limiting the depth of their findings. Here we present Kosmos, an AI scientist that automates data-driven discovery. Given an open-ended objective and a dataset, Kosmos runs for up to 12 hours performing cycles of parallel data analysis, literature search, and hypothesis generation before synthesizing discoveries into scientific reports. Unlike prior systems, Kosmos uses a structured world model to share information between a data analysis agent and a literature search agent. The world model enables Kosmos to coherently pursue the specified objective over 200 agent rollouts, collectively executing an average of 42,000 lines of code and reading 1,500 papers per run. Kosmos cites all statements in its reports with code or primary literature, ensuring its reasoning is traceable. Independent scientists found 79.4% of statements in Kosmos reports to be accurate, and collaborators reported that a single 20-cycle Kosmos run performed the equivalent of 6 months of their own research time on average. Furthermore, collaborators reported that the number of valuable scientific findings generated scales linearly with Kosmos cycles (tested up to 20 cycles). We highlight seven discoveries made by Kosmos that span metabolomics, materials science, neuroscience, and statistical genetics. Three discoveries independently reproduce findings from preprinted or unpublished manuscripts that were not accessed by Kosmos at runtime, while four make novel contributions to the scientific literature.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper presents Kosmos, an AI scientist system for autonomous data-driven discovery. Given an objective and dataset, it performs up to 12 hours of iterative cycles involving parallel data analysis, literature search, and hypothesis generation, using a structured world model to share information between agents and maintain coherence across up to 200 rollouts (averaging 42,000 lines of code and 1,500 papers read). All statements in generated reports are cited to code or primary literature. Key claims include 79.4% accuracy of statements as judged by independent scientists, a single 20-cycle run equating to 6 months of collaborator research time on average, linear scaling of valuable findings with cycles (tested to 20), and seven highlighted discoveries across metabolomics, materials science, neuroscience, and statistical genetics, including three that reproduce findings from preprinted or unpublished manuscripts not accessed at runtime.

Significance. If the empirical claims hold under rigorous validation, the work would mark a meaningful advance in AI systems for scientific discovery by demonstrating sustained multi-agent coherence over long horizons with built-in traceability. The structured world model for inter-agent information sharing and the citation mechanism are constructive design choices that address common failure modes in prior agents. The reproduction of unpublished results is a notable strength, as it provides a falsifiable test of generative capability. These elements, combined with reported linear scaling, could inform future autonomous discovery pipelines if the supporting evidence is strengthened.

major comments (3)
  1. [Abstract and Results] Abstract and Results (performance evaluation): The central claims of 79.4% statement accuracy and average 6-month research-time equivalence rest on external collaborator reports and independent review, yet the manuscript supplies no details on reviewer selection criteria, statement sampling procedure, or the quantification method for time equivalence. These gaps directly affect the soundness of the headline performance numbers.
  2. [Methods] Methods (structured world model and agent hand-offs): The assertion that the world model enables coherent objective pursuit across 200 rollouts by sharing information between data-analysis and literature agents is load-bearing for the scaling and depth claims, but no ablation (structured model vs. unstructured messaging), no objective-drift metric (e.g., fraction of hypotheses still aligned with the initial objective after N cycles), and no measurement of unsupported inferences are reported.
  3. [Results] Results (discovery validation): The seven highlighted discoveries, including three that reproduce findings from preprinted or unpublished manuscripts, are presented as evidence of value, but the manuscript does not describe verification protocols, controls against cherry-picking, or how these cases were selected from the full set of generated findings.
minor comments (2)
  1. [Abstract] The abstract states that collaborators reported linear scaling of valuable findings with cycles; a dedicated figure or table showing the per-cycle counts and any statistical test for linearity would improve clarity.
  2. [Introduction] Notation for agent rollouts and cycle counts is used without an explicit definition or diagram early in the manuscript; adding a concise schematic of the world-model update loop would aid readability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed review. The comments identify important areas for clarification that will improve the transparency and rigor of the manuscript. We address each major comment below and indicate the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract and Results] Abstract and Results (performance evaluation): The central claims of 79.4% statement accuracy and average 6-month research-time equivalence rest on external collaborator reports and independent review, yet the manuscript supplies no details on reviewer selection criteria, statement sampling procedure, or the quantification method for time equivalence. These gaps directly affect the soundness of the headline performance numbers.

    Authors: We agree that the evaluation details are insufficiently described. In the revised manuscript we will add a new subsection in Methods that specifies: reviewer selection (independent domain experts with no project involvement, recruited via academic networks); statement sampling (random selection of 50 statements per report from the full set of generated statements); and time-equivalence quantification (structured collaborator interviews in which researchers estimated hours required for equivalent manual tasks, averaged across five collaborators). These additions will allow readers to assess the reported figures directly. revision: yes

  2. Referee: [Methods] Methods (structured world model and agent hand-offs): The assertion that the world model enables coherent objective pursuit across 200 rollouts by sharing information between data-analysis and literature agents is load-bearing for the scaling and depth claims, but no ablation (structured model vs. unstructured messaging), no objective-drift metric (e.g., fraction of hypotheses still aligned with the initial objective after N cycles), and no measurement of unsupported inferences are reported.

    Authors: We acknowledge that the absence of an ablation and quantitative coherence metrics limits the strength of the claims. Due to the computational expense of full re-runs, we did not perform a complete ablation in the original experiments. In revision we will add a limited comparison using a subset of runs (structured vs. unstructured messaging) and report an objective-drift metric defined as the fraction of hypotheses retaining alignment with the initial objective after 5, 10, and 20 cycles, obtained via blinded annotation of a sample of 20 runs. We will also report the rate of unsupported inferences flagged by the same independent reviewers who assessed statement accuracy. revision: partial

  3. Referee: [Results] Results (discovery validation): The seven highlighted discoveries, including three that reproduce findings from preprinted or unpublished manuscripts, are presented as evidence of value, but the manuscript does not describe verification protocols, controls against cherry-picking, or how these cases were selected from the full set of generated findings.

    Authors: We selected the seven cases to illustrate both successful reproductions of unpublished work and novel contributions across domains. In the revision we will expand the Results section to describe the verification protocol (literature cross-check plus expert consultation for each finding) and the selection process (all reproducible or novel findings from the 20-cycle runs were eligible; the seven were chosen as representative examples rather than the complete set). A supplementary table listing all generated findings with verification status and selection rationale will be added to mitigate concerns about cherry-picking. revision: yes

Circularity Check

0 steps flagged

No significant circularity: empirical performance validated by external reviewers

full rationale

The paper reports system behavior and performance metrics (coherence over 200 rollouts, 79.4% statement accuracy, linear scaling of findings with cycles, and 6-month research equivalence) as outcomes of external collaborator reports and independent scientist reviews rather than as quantities derived from internal fitted parameters, self-referential definitions, or self-citation chains. The structured world model is presented as an architectural choice that enables information sharing, but its effectiveness is asserted via traceable citations and external judgment, not by reducing any claimed result to the model definition itself. No equations, predictions, or uniqueness theorems appear that collapse to the inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central performance claims rest on the assumption that the world model successfully maintains objective coherence and that the citation mechanism catches errors; no explicit free parameters or invented physical entities are described in the abstract.

axioms (1)
  • domain assumption The world model can be kept consistent across parallel data-analysis and literature agents for at least 200 rollouts without drift that invalidates downstream hypotheses.
    Invoked to explain why Kosmos can run far longer than prior agents before losing coherence.

pith-pipeline@v0.9.0 · 6004 in / 1362 out tokens · 33154 ms · 2026-05-22T08:36:08.188625+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 25 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Evaluating Large Language Models in Scientific Discovery

    cs.AI 2025-12 unverdicted novelty 8.0

    The SDE benchmark shows LLMs lag on scientific discovery tasks relative to general science tests, with diminishing scaling returns and shared weaknesses across models.

  2. Forecasting Scientific Progress with Artificial Intelligence

    cs.AI 2026-05 unverdicted novelty 7.0

    Introduces the CUSP benchmark across 4760 events and finds frontier AI models can pick plausible directions but fail to predict whether or when scientific advances will occur, with performance varying by domain and in...

  3. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

    cs.LG 2026-05 conditional novelty 7.0

    Starling uses LLMs and agents to turn 22.5M PubMed papers into 6.3M nuanced structured records across six tasks with 0.6-7.7% frontier-model rejection rates, lower than error rates on existing curated databases.

  4. AI co-mathematician: Accelerating mathematicians with agentic AI

    cs.AI 2026-05 unverdicted novelty 7.0

    An interactive AI workbench for mathematicians achieves 48% on FrontierMath Tier 4 and helped solve open problems in early tests.

  5. Optimizing ground state preparation protocols with autoresearch

    quant-ph 2026-04 unverdicted novelty 7.0

    AI coding agents evolve simple ground-state protocols into improved versions for VQE, DMRG, and AFQMC on spin models and molecules by using executable energy scores under fixed compute budgets.

  6. Optimizing ground state preparation protocols with autoresearch

    quant-ph 2026-04 unverdicted novelty 7.0

    AI coding agents mutate baseline protocols for VQE, DMRG, and AFQMC into versions with improved energy proxies on spin models and molecules while respecting computational budgets.

  7. AI scientists produce results without reasoning scientifically

    cs.AI 2026-04 conditional novelty 7.0

    LLM agents execute scientific tasks but fail to follow core scientific reasoning norms such as evidence consideration and belief revision based on refutations.

  8. CREATE: Testing LLMs for Associative Creativity

    cs.CL 2026-03 unverdicted novelty 7.0

    CREATE is a benchmark that scores LLMs on their ability to produce many specific and diverse associative paths between concepts drawn from parametric knowledge.

  9. El Agente Quntur: A research collaborator agent for quantum chemistry

    physics.chem-ph 2026-02 unverdicted novelty 7.0

    El Agente Quntur is a new multi-agent system that uses reasoning over literature and software documentation to autonomously handle the full workflow of quantum chemistry experiments in ORCA.

  10. Scalable Agentic Reasoning for Designing Biologics Targeting Intrinsically Disordered Proteins

    q-bio.QM 2025-12 unverdicted novelty 7.0

    StructBioReasoner is a scalable multi-agent system that designs IDP-targeting biologics, with over 50% of 787 candidates for Der f 21 showing better binding free energy than human-designed references.

  11. Towards Discovery of Polymers for Insulin Delivery via Physics-Grounded Agentic Workflows

    q-bio.QM 2026-05 unverdicted novelty 6.0

    An LLM-orchestrated physics simulation search identifies polymers with strong insulin interactions, outperforming standard optimization methods by significant margins.

  12. Unlocking LLM Creativity in Science through Analogical Reasoning

    cs.AI 2026-05 conditional novelty 6.0

    Analogical reasoning increases LLM solution diversity by 90-173% and novelty rate to over 50%, delivering up to 13-fold gains on biomedical tasks including perturbation prediction and cell communication.

  13. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

    cs.LG 2026-05 unverdicted novelty 6.0

    An LLM entity-tagging pipeline plus multi-agent system extracts ~6.3M nuanced records from 22.5M PubMed papers across six tasks with lower measured error than existing curated databases.

  14. AI co-mathematician: Accelerating mathematicians with agentic AI

    cs.AI 2026-05 unverdicted novelty 6.0

    An interactive AI workbench called the AI co-mathematician supports open-ended mathematical research and achieves a new high score of 48% on FrontierMath Tier 4.

  15. Intentmaking and Sensemaking: Human Interaction with AI-Guided Mathematical Discovery

    cs.AI 2026-05 unverdicted novelty 6.0

    Expert mathematicians using an AI coding agent for discovery engage in repeated cycles of intentmaking to define goals and sensemaking to interpret outputs.

  16. Hypothesis generation and updating in large language models

    cs.LG 2026-05 unverdicted novelty 6.0

    LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.

  17. SciResearcher: Scaling Deep Research Agents for Frontier Scientific Reasoning

    cs.AI 2026-05 unverdicted novelty 6.0

    SciResearcher automates creation of diverse scientific reasoning tasks from academic evidence to train an 8B model that sets new SOTA at 19.46% on HLE-Bio/Chem-Gold and gains 13-15% on SuperGPQA-Hard-Biology and TRQA-...

  18. PRL-Bench: A Comprehensive Benchmark Evaluating LLMs' Capabilities in Frontier Physics Research

    cs.LG 2026-04 unverdicted novelty 6.0

    PRL-Bench evaluates frontier LLMs on 100 real physics research tasks and finds the best models score below 50, exposing a gap to autonomous discovery.

  19. DeepReviewer 2.0: A Traceable Agentic System for Auditable Scientific Peer Review

    cs.AI 2026-03 unverdicted novelty 6.0

    An agentic system produces traceable review packages and an un-finetuned 196B model using it covers more major issues than Gemini-3.1-Pro on 134 ICLR 2025 submissions while winning most blind comparisons to human committees.

  20. Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task

    cs.CL 2026-02 unverdicted novelty 6.0

    LLMs diverge from human goal selection in self-directed learning by exploiting single solutions with low variability across instances.

  21. CodeDistiller: Automatically Generating Code Libraries for Scientific Coding Agents

    cs.AI 2025-11 conditional novelty 6.0

    CodeDistiller distills 250 materials-science GitHub repositories into vetted code libraries that improve the accuracy and scientific soundness of experiments generated by ASD agents.

  22. Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators

    cs.MA 2026-05 unverdicted novelty 5.0

    Sibyl-AutoResearch introduces self-evolving trial-and-error harnesses with auditable conversion units that link trial signals to updated research behaviors and harness repairs in autonomous systems.

  23. Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks

    cs.AI 2026-05 unverdicted novelty 5.0

    A multi-agent harness autonomously generates functional single-page VIS apps with linked views for scientific data tasks using coordinated skills for analysis, planning, implementation, and evaluation.

  24. AI for Auto-Research: Roadmap & User Guide

    cs.AI 2026-05 unverdicted novelty 4.0

    The paper delivers a stage-by-stage roadmap for AI in research, showing reliable assistance in retrieval and tool tasks but fragility in novelty and judgment, advocating human-governed collaboration.

  25. Are Researchers Being Replaced by Artificial Intelligence?

    cs.CY 2026-04 unverdicted novelty 3.0

    AI is shifting researchers from creators to curators of generated content, risking loss of intellectual ownership and genuine understanding of science.

Reference graph

Works this paper leans on

70 extracted references · 70 canonical work pages · cited by 22 Pith papers · 2 internal anchors

  1. [1]

    Robin: A multi-agent system for automating scientific discovery

    Ali Essam Ghareeb, Benjamin Chang, Ludovico Mitchener, Angela Yiu, Caralyn J. Szostkiewicz, Jon M. Laurent, Muhammed T. Razzak, Andrew D. White, Michaela M. Hinks, and Samuel G. Rodriques. Robin: A multi-agent system for automating scientific discovery, May 2025. arXiv:2505.13400 [cs]

    work page arXiv 2025

  2. [2]

    The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery

    Chris Lu, Cong Lu, Robert Tjarko Lange, Jakob Foerster, Jeff Clune, and David Ha. The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery, September 2024. arXiv:2408.06292 [cs]

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  3. [3]

    Juraj Gottweis, Wei-Hung Weng, Alexander Daryin, Tao Tu, Anil Palepu, Petar Sirkovic, Artiom Myaskovsky, Fe- lix Weissenberger, Keran Rong, Ryutaro Tanno, Khaled Saab, Dan Popovici, Jacob Blum, Fan Zhang, Katherine Chou, Avinatan Hassidim, Burak Gokturk, Amin Vahdat, Pushmeet Kohli, Yossi Matias, Andrew Carroll, Kavita Kulkarni, Nenad Tomasev, Yuan Guan, ...

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  4. [4]

    Bulaong, John E

    Kyle Swanson, Wesley Wu, Nash L. Bulaong, John E. Pak, and James Zou. The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies.Nature, 646(8085):716–723, October 2025. Publisher: Nature Publishing Group

    work page 2025

  5. [5]

    Laurent, Alex Andonian, Benjamin Tenmann, Siddharth Narayanan, Geemi P

    Ludovico Mitchener, Jon M. Laurent, Alex Andonian, Benjamin Tenmann, Siddharth Narayanan, Geemi P. Wellawatte, Andrew White, Lorenzo Sani, and Samuel G. Rodriques. BixBench: a Comprehensive Benchmark for LLM-based Agents in Computational Biology, October 2025. arXiv:2503.00096 [q-bio]

    work page arXiv 2025

  6. [6]

    Skarlinski, Sam Cox, Jon M

    Michael D. Skarlinski, Sam Cox, Jon M. Laurent, James D. Braza, Michaela Hinks, Michael J. Hammerling, Manvitha Ponnapati, Samuel G. Rodriques, and Andrew D. White. Language agents achieve superhuman synthesis of scientific knowledge, September 2024. arXiv:2409.13740 [cs]

    work page arXiv 2024

  7. [7]

    Yunheng Zou, Austin H. Cheng, Abdulrahman Aldossary, Jiaru Bai, Shi Xuan Leong, Jorge Arturo Campos- Gonzalez-Angulo, Changhyeok Choi, Cher Tian Ser, Gary Tom, Andrew Wang, Zijian Zhang, Ilya Yakavets, Han Hao, Chris Crebolder, Varinia Bernales, and Alán Aspuru-Guzik. El Agente: An autonomous agent for quantum chemistry.Matter, 8(7), July 2025. Publisher:...

    work page 2025

  8. [8]

    Ziegler, Elizabeth Barnes, and Lawrence Chan

    Thomas Kwa, Ben West, Joel Becker, Amy Deng, Katharyn Garcia, Max Hasin, Sami Jawhar, Megan Kinniment, Nate Rush, Sydney Von Arx, Ryan Bloom, Thomas Broadley, Haoxing Du, Brian Goodrich, Nikola Jurkovic, Luke Harold Miles, Seraphina Nix, Tao Lin, Neev Parikh, David Rein, Lucas Jun Koba Sato, Hjalmar Wijk, Daniel M. Ziegler, Elizabeth Barnes, and Lawrence ...

    work page

  9. [9]

    arXiv:2503.14499 [cs]

    work page arXiv

  10. [10]

    Gonzales, Khairunisa Mohamad Ibrahim, Fan Zhang, Hannah E

    Aizad Kamal, Juan Liu, Ernesto R. Gonzales, Khairunisa Mohamad Ibrahim, Fan Zhang, Hannah E. Skelton, Javier Kelly Cuenca, Carla Yuede, Gary J. Patti, Leah P. Shriver, Jin-Moo Lee, Aaron J. Norris, and Eric C. Land- sness. Preoptic activation induces a torpor-like hypothermic and hypometabolic state that is cerebroprotective, October 2025. ISSN: 2692-8205...

    work page 2025

  11. [11]

    PyrimidineSalvage: Physiological Functions and Interaction with Chloroplast Biogenesis.Plant Physiology, 180(4):1816–1828, August 2019

    LisaOhler, SandraNiopek-Witz, SamuelE.Mainguet, andTorstenMöhlmann. PyrimidineSalvage: Physiological Functions and Interaction with Chloroplast Biogenesis.Plant Physiology, 180(4):1816–1828, August 2019

    work page 2019

  12. [12]

    Purine nucleosides: endogenous neuro- protectants in hypoxic brain.Journal of Neurochemistry, 121(3):329–342, May 2012

    Bettina Thauerer, Stephanie zur Nedden, and Gabriele Baier-Bitterlich. Purine nucleosides: endogenous neuro- protectants in hypoxic brain.Journal of Neurochemistry, 121(3):329–342, May 2012

    work page 2012

  13. [13]

    Harvey, Paula C

    Hui Shen, Guann-Juh Chen, Brandon K. Harvey, Paula C. Bickford, and Yun Wang. Inosine reduces ischemic brain injury in rats.Stroke, 36(3):654–659, March 2005

    work page 2005

  14. [14]

    L. I. Benowitz, D. E. Goldberg, J. R. Madsen, D. Soni, and N. Irwin. Inosine stimulates extensive axon collateral growth in the rat corticospinal tract after injury.Proceedings of the National Academy of Sciences of the United States of America, 96(23):13486–13490, November 1999

    work page 1999

  15. [15]

    Formica, Moungi G

    Tianran Liu, Nicky Evans, Kangyu Ji, Ronaldo Lee, Aaron Zhu, Vinn Nguyen, James Serdy, Elizabeth Wall, Yongli Lu, Florian A. Formica, Moungi G. Bawendi, Quinn C. Burlingame, Yueh-Lin Loo, Vladimir Bulovic, and Tonio Buonassisi. Disentangling the Effects of Simultaneous Environmental Variables on Perovskite Synthesis and Device Performance via Interpretabl...

    work page arXiv 2025

  16. [16]

    Barabási, André Ferreira Castro, Giulia Menichetti, and Albert-László Barabási

    Ben Piazza, Dániel L. Barabási, André Ferreira Castro, Giulia Menichetti, and Albert-László Barabási. Physical Network Constraints Define the Lognormal Architecture of the Brain’s Connectome, February 2025. Pages: 2025.02.27.640551 Section: New Results

    work page 2025

  17. [17]

    Problems with classification, hypothesis testing, and estimator convergence in the analysis of degree distributions in networks, March 2020

    Pim van der Hoorn, Ivan Voitalov, Remco van der Hofstad, and Dmitri Krioukov. Problems with classification, hypothesis testing, and estimator convergence in the analysis of degree distributions in networks, March 2020. arXiv:2003.14012 [physics]

    work page arXiv 2020

  18. [18]

    Tschumperlin, and Naveen L

    Rachad Ghazal, Min Wang, Duan Liu, Daniel J. Tschumperlin, and Naveen L. Pereira. Cardiac Fibrosis in the Multi-Omics Era: Implications for Heart Failure.Circulation Research, 136(7):773–802, March 2025

    work page 2025

  19. [19]

    MicroRNAs regulating superoxide dismutase 2 are new circulating biomarkers of heart failure.Scientific Reports, 7(1):14747, November 2017

    Emilie Dubois-Deruy, Marie Cuvelliez, Jan Fiedler, Henri Charrier, Paul Mulder, Eleonore Hebbar, Angelika Pfanne, Olivia Beseme, Maggy Chwastyniak, Philippe Amouyel, Vincent Richard, Christophe Bauters, Thomas Thum, and Florence Pinet. MicroRNAs regulating superoxide dismutase 2 are new circulating biomarkers of heart failure.Scientific Reports, 7(1):1474...

    work page 2017

  20. [20]

    Arendshorst, Aleksandr E

    Willaim J. Arendshorst, Aleksandr E. Vendrov, Nitin Kumar, Santhi K. Ganesh, and Nageswara R. Madamanchi. Oxidative Stress in Kidney Injury and Hypertension.Antioxidants, 13(12):1454, December 2024. Publisher: Multidisciplinary Digital Publishing Institute

    work page 2024

  21. [21]

    Superoxide Dismutases: Role in Redox Signaling, Vascular Function, and Diseases.Antioxidants & Redox Signaling, 15(6):1583–1606, September 2011

    Tohru Fukai and Masuko Ushio-Fukai. Superoxide Dismutases: Role in Redox Signaling, Vascular Function, and Diseases.Antioxidants & Redox Signaling, 15(6):1583–1606, September 2011. Publisher: Mary Ann Liebert, Inc., publishers

    work page 2011

  22. [22]

    The Role of Endothelial Superoxide Dismutase 2 in Modulating Sickle Pathology.Blood, 138(Supplement 1):190, November 2021

    Atinuke Dosunmu-Ogunbi, Shuai Yuan, and Adam C Straub. The Role of Endothelial Superoxide Dismutase 2 in Modulating Sickle Pathology.Blood, 138(Supplement 1):190, November 2021

    work page 2021

  23. [23]

    St Clair, Md Shenuarin Bhuiyan, Christo- pher Kevil, Megan N

    Sudha Sharma, Susmita Bhattarai, Hosne Ara, Grace Sun, Daret K. St Clair, Md Shenuarin Bhuiyan, Christo- pher Kevil, Megan N. Watts, Paari Dominic, Takahiko Shimizu, Kevin J. McCarthy, Hong Sun, Manikandan Panchatcharam, and Sumitra Miriyala. SOD2 deficiency in cardiomyocytes defines defective mitochondrial bioenergetics as a cause of lethal dilated cardi...

    work page 2020

  24. [24]

    Kliment, Hagir B

    Corrine R. Kliment, Hagir B. Suliman, Jacob M. Tobolewski, Crystal M. Reynolds, Brian J. Day, Xiaodong Zhu, Charles F. McTiernan, Kenneth R. McGaffin, Claude A. Piantadosi, and Tim D. Oury. Extracellular superoxide dismutase regulates cardiac function and fibrosis.Journal of Molecular and Cellular Cardiology, 47(5):730–742, November 2009. Publisher: Elsevier

    work page 2009

  25. [25]

    Xiu-Yun Jiang, Qing Chen, Xiao-Yu Chen, Qiu-Ying Sun, Fei Jing, Hai-Qing Zhang, Jin Xu, Xiao-Hong Li, and Qing-Bo Guan. Superoxide dismutases: marker in predicting reduced left ventricular ejection fraction in patients with type 2 diabetes and acute coronary syndrome.BMC Cardiovascular Disorders, 24(1):191, April 2024

    work page 2024

  26. [26]

    Victor Nauffal, Paolo Di Achille, Marcus D. R. Klarqvist, Jonathan W. Cunningham, Matthew C. Hill, James P. Pirruccello, Lu-Chen Weng, Valerie N. Morrill, Seung Hoan Choi, Shaan Khurshid, Samuel F. Friedman, Mahan Nekoui, Carolina Roselli, Kenney Ng, Anthony A. Philippakis, Puneet Batra, Patrick T. Ellinor, and Steven A. Lubitz. Genetics of myocardial int...

    work page 2023

  27. [27]

    Keaton, Julie A

    Marijana Vujkovic, Jacob M. Keaton, Julie A. Lynch, Donald R. Miller, Jin Zhou, Catherine Tcheandjieu, Jennifer E. Huffman, Themistocles L. Assimes, Kim Lorenz, Xiang Zhu, Austin T. Hilliard, Renae L. Judy, Jie Huang, Kyung M. Lee, Derek Klarin, Saiju Pyarajan, John Danesh, Olle Melander, Asif Rasheed, Nadeem H. Mallick, Shahid Hameed, Irshad H. Qureshi, ...

    work page 2020

  28. [28]

    Spracklen, Weihua Zhang, Maggie C

    Anubha Mahajan, Cassandra N. Spracklen, Weihua Zhang, Maggie C. Y. Ng, Lauren E Petty, Hidetoshi Kitajima, Grace Z. Yu, Sina Rüeger, Leo Speidel, Young Jin Kim, Momoko Horikoshi, Josep M. Mercader, Daniel Taliun, Sanghoon Moon, Soo-Heon Kwak, Neil R. Robertson, Nigel W. Rayner, Marie Loh, Bong-Jo Kim, Joshua Chiou, Irene Miguel-Escalada, Pietro della Brio...

    work page 2022

  29. [29]

    Kim, Patrick E

    Gaowei Wang, Joshua Chiou, Chun Zeng, Michael Miller, Ileana Matta, Jee Yun Han, Nikita Kadakia, Mei- Lin Okino, Elisha Beebe, Medhavi Mallick, Joan Camunas-Soler, Theodore dos Santos, Xiao-Qing Dai, Cara Ellis, Yan Hang, Seung K. Kim, Patrick E. MacDonald, Fouad R. Kandeel, Sebastian Preissl, Kyle J. Gaulton, and Maike Sander. Integrating genetics with s...

    work page 2023

  30. [30]

    The GTEx Consortium atlas of genetic regulatory effects across human tissues.Science, 369(6509):1318–1330, September 2020

    The GTEx Consortium. The GTEx Consortium atlas of genetic regulatory effects across human tissues.Science, 369(6509):1318–1330, September 2020. Publisher: American Association for the Advancement of Science

    work page 2020

  31. [31]

    Burren, Jonathan Davitte, Kyle L

    BenjaminB.Sun, JoshuaChiou, MatthewTraylor, ChristianBenner, Yi-HsiangHsu, TomG.Richardson, Praveen Surendran, AnubhaMahajan, ChloeRobins, StevenG.Vasquez-Grinnell, LipingHou, ErikaM.Kvikstad, OliverS. Burren, Jonathan Davitte, Kyle L. Ferber, Christopher E. Gillies, Åsa K. Hedman, Sile Hu, Tinchi Lin, Rajesh Mikkilineni, Rion K. Pendergrass, Corran Picke...

    work page

  32. [33]

    motifbreakR v2: expanded variant analysis including indels and integrated evidence from transcription factor binding databases.Bioinformatics Advances, 4(1):vbae162, January 2024

    Simon G Coetzee and Dennis J Hazelett. motifbreakR v2: expanded variant analysis including indels and integrated evidence from transcription factor binding databases.Bioinformatics Advances, 4(1):vbae162, January 2024

    work page 2024

  33. [34]

    Gschwind, Kristy S

    Andreas R. Gschwind, Kristy S. Mualim, Alireza Karbalayghareh, Maya U. Sheth, Kushal K. Dey, Evelyn Jagoda, Ramil N. Nurtdinov, Wang Xi, Anthony S. Tan, Hank Jones, X. Rosa Ma, David Yao, Joseph Nasser, Žiga Avsec, Benjamin T. James, Muhammad S. Shamim, Neva C. Durand, Suhas S. P. Rao, Ragini Mahajan, Benjamin R. Doughty, Kalina Andreeva, Jacob C. Ulirsch...

    work page 2023

  34. [35]

    Alexander Gusev, Arthur Ko, Huwenbo Shi, Gaurav Bhatia, Wonil Chung, Brenda W. J. H. Penninx, Rick Jansen, Eco J. C. de Geus, Dorret I. Boomsma, Fred A. Wright, Patrick F. Sullivan, Elina Nikkola, Marcus Alvarez, Mete Civelek, Aldons J. Lusis, Terho Lehtimäki, Emma Raitoharju, Mika Kähönen, Ilkka Seppälä, Olli T. Raitakari, Johanna Kuusisto, Markku Laakso...

    work page 2016

  35. [36]

    Deficient endoplasmic reticulum translocon-associated protein complex limits the biosynthesis of proinsulin and insulin.The FASEB Journal, 35(5):e21515, 2021

    Yumeng Huang, Xiaoxi Xu, Peter Arvan, and Ming Liu. Deficient endoplasmic reticulum translocon-associated protein complex limits the biosynthesis of proinsulin and insulin.The FASEB Journal, 35(5):e21515, 2021. _eprint: https://faseb.onlinelibrary.wiley.com/doi/pdf/10.1096/fj.202002774R

    work page doi:10.1096/fj.202002774r 2021

  36. [37]

    Trapαdeficiency impairs the early events of insulin biosynthesis and glucose homeostasis.The Journal of clinical investigation, 135(14):e179845, July 2025

    Xin Li, Jingxin Hu, Yumeng Huang, Hai Zhang, Ning Xu, Yang Liu, Xuan Liu, Yuanyuan Ye, Xinxin Zhang, Xiaoxi Xu, Yuxin Fan, Ziyue Zhang, Weiping J Zhang, Shusen Wang, Wenli Feng, Peter Arvan, and Ming Liu. Trapαdeficiency impairs the early events of insulin biosynthesis and glucose homeostasis.The Journal of clinical investigation, 135(14):e179845, July 2025

    work page 2025

  37. [38]

    Tau Protein Modulates Perineuronal Extracellular Matrix Expression in the TauP301L-acan Mouse Model.Biomolecules, 12(4):505, April 2022

    Sophie Schmidt, Max Holzer, Thomas Arendt, Mandy Sonntag, and Markus Morawski. Tau Protein Modulates Perineuronal Extracellular Matrix Expression in the TauP301L-acan Mouse Model.Biomolecules, 12(4):505, April 2022. Publisher: Multidisciplinary Digital Publishing Institute. 30

    work page 2022

  38. [39]

    Mahajani, Debia Wakhloo, Weijing Tang, Yue-Qiang Xue, Samuel Morabito, Jie Pan, Jane Oberhauser, Angela E

    Marcos Otero-Garcia, Sameehan U. Mahajani, Debia Wakhloo, Weijing Tang, Yue-Qiang Xue, Samuel Morabito, Jie Pan, Jane Oberhauser, Angela E. Madira, Tamara Shakouri, Yongning Deng, Thomas Allison, Zihuai He, William E. Lowry, Riki Kawaguchi, Vivek Swarup, and Inma Cobos. Molecular signatures underlying neurofibril- lary tangle susceptibility in Alzheimer’s...

    work page 2022

  39. [40]

    A. K. Wagner, S. B. Soumerai, F. Zhang, and D. Ross-Degnan. Segmented regression analysis of interrupted time series studies in medication use research.Journal of Clinical Pharmacy and Therapeutics, 27(4):299–309, August 2002

    work page 2002

  40. [41]

    Fox, Syed A

    Dunja Mrdjen, Edward J. Fox, Syed A. Bukhari, Kathleen S. Montine, Sean C. Bendall, and Thomas J. Montine. The basis of cellular and regional vulnerability in Alzheimer’s disease.Acta Neuropathologica, 138(5):729–749, November 2019

    work page 2019

  41. [42]

    Naoto Watamura, Martha S. Foiani, Sumi Bez, Mathieu Bourdenx, Alessia Santambrogio, Claire Frodsham, Elena Camporesi, Gunnar Brinkmalm, Henrik Zetterberg, Saisha Patel, Naoko Kamano, Mika Takahashi, Javier Rueda-Carrasco, Loukia Katsouri, Stephanie Fowler, Emir Turkes, Shoko Hashimoto, Hiroki Sasaguri, Takashi Saito, AFM Saiful Islam, Seico Benner, Toshih...

    work page 2025

  42. [43]

    Hongjun Fu, John Hardy, and Karen E. Duff. Selective vulnerability in neurodegenerative diseases.Nature Neuroscience, 21(10):1350–1358, October 2018. Publisher: Nature Publishing Group

    work page 2018

  43. [44]

    Local externalization of phosphatidylserine mediates developmental synaptic pruning by microglia.The EMBO Journal, 39(16):e105380, August 2020

    Nicole Scott-Hewitt, Fabio Perrucci, Raffaella Morini, Marco Erreni, Matthew Mahoney, Agata Witkowska, Alanna Carey, Elisa Faggiani, Lisa Theresia Schuetz, Sydney Mason, Matteo Tamborini, Matteo Bizzotto, Lorena Passoni, Fabia Filipello, Reinhard Jahn, Beth Stevens, and Michela Matteoli. Local externalization of phosphatidylserine mediates developmental s...

    work page 2020

  44. [45]

    Kelly Jin, Zizhen Yao, Cindy T. J. van Velthoven, Eitan S. Kaplan, Katie Glattfelder, Samuel T. Barlow, Gabriella Boyer, Daniel Carey, Tamara Casper, Anish Bhaswanth Chakka, Rushil Chakrabarty, Michael Clark, Max Departee, Marie Desierto, Amanda Gary, Jessica Gloe, Jeff Goldy, Nathan Guilford, Junitta Guzman, Daniel Hirschstein, Changkyu Lee, Elizabeth Li...

    work page

  45. [47]

    Ehrenberg, Car- los A

    Kun Leng, Emmy Li, Rana Eser, Antonia Piergies, Rene Sit, Michelle Tan, Norma Neff, Song Hua Li, Roberta Diehl Rodriguez, Claudia Kimie Suemoto, Renata Elaine Paraizo Leite, Alexander J. Ehrenberg, Car- los A. Pasqualucci, William W. Seeley, Salvatore Spina, Helmut Heinsen, Lea T. Grinberg, and Martin Kamp- mann. Molecular characterization of selectively ...

    work page 2021

  46. [48]

    Microglia-synapse engulfment via PtdSer-TREM2 ameliorates neuronal hyperactivity in Alzheimer’s disease models.The EMBO Journal, 42(19):e113246, October 2023

    Javier Rueda-Carrasco, Dimitra Sokolova, Sang-Eun Lee, Thomas Childs, Natália Jurčáková, Gerard Crowley, Sebastiaan De Schepper, Judy Z Ge, Joanne I Lachica, Christina E Toomey, Oliver J Freeman, John Hardy, Samuel J Barnes, Tammaryn Lashley, Beth Stevens, Sunghoe Chang, and Soyon Hong. Microglia-synapse engulfment via PtdSer-TREM2 ameliorates neuronal hy...

    work page 2023

  47. [49]

    Regulation of phospholipid distribution in the lipid bilayer by flippases and scramblases.Nature Reviews Molecular Cell Biology, 24(8):576–596, August 2023

    Takaharu Sakuragi and Shigekazu Nagata. Regulation of phospholipid distribution in the lipid bilayer by flippases and scramblases.Nature Reviews Molecular Cell Biology, 24(8):576–596, August 2023. Publisher: Nature Publishing Group

    work page 2023

  48. [50]

    Stoub, Leyla deToledo Morrell, Glenn T

    Travis R. Stoub, Leyla deToledo Morrell, Glenn T. Stebbins, Sue Leurgans, David A. Bennett, and Raj C. Shah. Hippocampal disconnection contributes to memory dysfunction in individuals at risk for Alzheimer’s disease. Proceedings of the National Academy of Sciences, 103(26):10041–10045, June 2006. Publisher: Proceedings of the National Academy of Sciences. 31

    work page 2006

  49. [51]

    J. G. White, E. Southgate, J. N. Thomson, and S. Brenner. The structure of the nervous system of the nematode Caenorhabditis elegans.Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 314(1165):1–340, November 1986

    work page 1986

  50. [52]

    Pedigo, Christopher L

    Michael Winding, Benjamin D. Pedigo, Christopher L. Barnes, Heather G. Patsolic, Youngser Park, Tom Kaz- imiers, Akira Fushiki, Ingrid V. Andrade, Avinash Khandelwal, Javier Valdes-Aleman, Feng Li, Nadine Randel, Elizabeth Barsotti, Ana Correia, Richard D. Fetter, Volker Hartenstein, Carey E. Priebe, Joshua T. Vogel- stein, Albert Cardona, and Marta Zlati...

    work page 2023

  51. [53]

    Sterling, Philipp Schlegel, Szi-chieh Yu, Claire E

    Sven Dorkenwald, Arie Matsliah, Amy R. Sterling, Philipp Schlegel, Szi-chieh Yu, Claire E. McKellar, Albert Lin, Marta Costa, Katharina Eichler, Yijie Yin, Will Silversmith, Casey Schneider-Mizell, Chris S. Jordan, Derrick Brittain, Akhilesh Halageri, Kai Kuehner, Oluwaseun Ogedengbe, Ryan Morey, Jay Gager, Krzysztof Kruk, Eric Perlman, Runzhe Yang, David...

    work page 2024

  52. [54]

    A connectome and analysis of the adult Drosophila central brain.eLife, 9:e57443, September 2020

    Louis K Scheffer, C Shan Xu, Michal Januszewski, Zhiyuan Lu, Shin-ya Takemura, Kenneth J Hayworth, Gary B Huang, Kazunori Shinomiya, Jeremy Maitlin-Shepard, Stuart Berg, Jody Clements, Philip M Hubbard, William T Katz, Lowell Umayam, Ting Zhao, David Ackerman, Tim Blakely, John Bogovic, Tom Dolafi, Dagmar Kain- mueller, Takashi Kawase, Khaled A Khairy, La...

    work page 2020

  53. [55]

    Hayworth, Gary B

    Shin-ya Takemura, Kenneth J. Hayworth, Gary B. Huang, Michal Januszewski, Zhiyuan Lu, Elizabeth C. Marin, Stephan Preibisch, C. Shan Xu, John Bogovic, Andrew S. Champion, Han SJ Cheong, Marta Costa, Katharina Eichler, WilliamKatz, ChristopherKnecht, FengLi, BillyJ.Morris, ChristopherOrdish, PatriciaK.Rivlin, Philipp Schlegel, Kazunori Shinomiya, Tomke Stü...

    work page 2024

  54. [56]

    Ramirez, Runzhe Yang, Nico Kemnitz, Dodam Ih, Nicholas Turner, Kisuk Lee, Ignacio Tartavull, William M

    Ashwin Vishwanathan, Alex Sood, Jingpeng Wu, Alexandro D. Ramirez, Runzhe Yang, Nico Kemnitz, Dodam Ih, Nicholas Turner, Kisuk Lee, Ignacio Tartavull, William M. Silversmith, Chris S. Jordan, Celia David, Doug Bland, Amy Sterling, H. Sebastian Seung, Mark S. Goldman, and Emre R. F. Aksay. Predicting modular functions and 32 neural coding of behavior from ...

    work page

  55. [57]

    Publisher: Nature Publishing Group

    work page

  56. [58]

    Schneider-Mizell, Agnes L

    Casey M. Schneider-Mizell, Agnes L. Bodor, Derrick Brittain, JoAnn Buchanan, Daniel J. Bumbarger, Leila Elabbady, Clare Gamlin, Daniel Kapner, Sam Kinn, Gayathri Mahalingam, Sharmishtaa Seshamani, Shelby Suckow, Marc Takeno, Russel Torres, Wenjing Yin, Sven Dorkenwald, J. Alexander Bae, Manuel A. Castro, Akhilesh Halageri, Zhen Jia, Chris Jordan, Nico Kem...

    work page 2025

  57. [59]

    Berger, Art Pope, Yuelong Wu, Tim Blakely, Richard L

    Alexander Shapson-Coe, Michał Januszewski, Daniel R. Berger, Art Pope, Yuelong Wu, Tim Blakely, Richard L. Schalek, Peter H. Li, Shuohong Wang, Jeremy Maitin-Shepard, Neha Karlupia, Sven Dorkenwald, Evelina Sjost- edt, Laramie Leavitt, Dongil Lee, Jakob Troidl, Forrest Collman, Luke Bailey, Angerica Fitzmaurice, Rohin Kar, Benjamin Field, Hank Wu, Julian ...

    work page 2024

  58. [60]

    Yavorska and Stephen Burgess

    Olena O. Yavorska and Stephen Burgess. MendelianRandomization: an R package for performing Mendelian ran- domization analyses using summarized data.International Journal of Epidemiology, 46(6):1734–1739, December 2017

    work page 2017

  59. [61]

    Characterization of cell fate probabilities in single-cell data with Palantir.Nature Biotechnology, 37(4):451–460, April 2019

    Manu Setty, Vaidotas Kiseliovas, Jacob Levine, Adam Gayoso, Linas Mazutis, and Dana Pe’er. Characterization of cell fate probabilities in single-cell data with Palantir.Nature Biotechnology, 37(4):451–460, April 2019

    work page 2019

  60. [62]

    Zizhen Yao, Cindy T. J. van Velthoven, Michael Kunst, Meng Zhang, Delissa McMillen, Changkyu Lee, Won Jung, Jeff Goldy, Aliya Abdelhak, Matthew Aitken, Katherine Baker, Pamela Baker, Eliza Barkan, Darren Bertagnolli, Ashwin Bhandiwad, Cameron Bielstein, Prajal Bishwakarma, Jazmin Campos, Daniel Carey, Tamara Casper, Anish Bhaswanth Chakka, Rushil Chakraba...

    work page 2023

  61. [63]

    Describe the specific claim that you previously made

    Consider a claim you previously made similar in style and scope to the one from which Kosmos reproduced the discovery. Describe the specific claim that you previously made

    work page

  62. [64]

    Please review the following Kosmos report

    How long did it take to produce that claim after you had collected the data? Please answer in terms of elapsed calendar time (days/weeks/months/years). Please review the following Kosmos report. Note that you can click any hyperlinked trajectory to review them

    work page

  63. [65]

    Please list out the number of all findings in this report that you found to be valuable or significant

    work page

  64. [66]

    Assume your lab received the data on January 1. Under your usual workflow without AI tools, how long would it typically take your lab to generate findings like those in the report, including time spent reading the literature, coming up with ideas, performing the analyses? Please answer in terms of elapsed calendar time in terms of days/weeks/months/years

    work page

  65. [67]

    How many full-time employees would you expect to be working on the project during that period? For example, 1, 1.25, 1.7 FTE

    work page

  66. [68]

    Assess the depth of those findings that you chose, in terms of whether they required deep multi-step reasoning or more shallow ideation: •Shallow ideation based on surface-level observations •Light reasoning with a straightforward connection of ideas/observations •Moderate depth of reasoning involving several connected steps •Deep multi-step reasoning req...

    work page

  67. [69]

    Assess the quality of those findings that you chose in terms of novelty to the field: •Little novel insight, mostly restatement of known information •Moderate novelty; extends existing knowledge in meaningful ways •Largely novel with minor connections to existing knowledge •Completely novel insight with little precedent 38 Supplementary Information 3: Exa...

    work page

  68. [70]

    This report will provide context for the statements that you will be evaluating

    Read through the PDF of the discovery group. This report will provide context for the statements that you will be evaluating. The figures are there for your reference, but they will not be evaluated. Do not look through the hyperlinked citations or use those as evidence

    work page

  69. [71]

    All analyses and discoveries in this discovery group were derived from this dataset

    Familiarize yourself with the dataset provided. All analyses and discoveries in this discovery group were derived from this dataset

    work page

  70. [72]

    specialized

    Each discovery group will have a sample of statements to be evaluated. These claims are either derived from data analysis on the dataset, literature review, or by synthesizing ideas across claims/conclusions. Conduct your evaluation based on the type of claim it is. We tried our best to make each statement individually evaluable, but if something is confu...

    work page doi:10.1371/journal.pmed.1001871