Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2611.05413

Morpheus.AI: agents autonomously generate, execute and refine virtual tissue models

Prerana Chandratre·Anjali Sharma·Justin Bürger Michaelis·Diego Jahn·Jörn Starruß·Lutz Brusch·Simon Syga
View original source ↗Read source document ↗Export metadata ↓

Abstract

Virtual tissue models link observations to mechanistic hypotheses, but building them demands expertise in biology, mathematics and programming. Here, large language model (LLM) agents paired with the simulator Morpheus (Morpheus.AI) autonomously built such models from fifteen benchmark studies, five without a published model. With a frontier LLM, agents that ran their simulations met 83-86% of study-specific criteria in under two hours and for under US\$4 per model. Such agents could make in silico modeling routine.

Explore related subjects

q-bio.QMq-bio.CBq-bio.TO

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Prerana Chandratre, Anjali Sharma, Justin Bürger Michaelis, Diego Jahn, Jörn Starruß, Lutz Brusch, Simon Syga. 2026-10-29. Morpheus.AI: agents autonomously generate, execute and refine virtual tissue models. https://arxiv.org/abs/2611.05413

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Molecular notation can act as a shortcut in AI benchmark evaluation

High predictive accuracy does not necessarily mean that an AI model relies on information relevant to its scientific task. Molecular prediction illustrates this concern: the same molecule can be represented by different SMILES strings, potentially altering model behavior. We combined a systematic PubMed literature survey with experiments on chemical language models. SMILES preparation could not be identified from articles or Supplementary Information in 71 of 174 studies (40.8%). Across six benchmarks, 58.7% of identity-preserving input pairs used different SMILES strings. Notation differences changed reconstruction and latent representations, while separately trained classifiers often achieved similar scores. In two datasets, notation was associated with labels and changed predictions of a fixed model, showing how benchmark scores can reward notational shortcuts. Balancing notation during classifier training reduced prediction differences but sometimes lowered performance on original inputs. These findings support evaluating predictive performance alongside consistency under identity-preserving changes in input representation.

q-bio.QM↗
arXiv2026

A Comprehensive Benchmark of Computational Models for Antibody--Antigen Binding Affinity

Computational antibody design requires benchmarks that measure functional binding rather than similarity to natural sequences. Here we present AbBiBench, which treats the antibody--antigen complex as the evaluation unit and harmonizes 216,560 experimental measurements spanning 13 antibodies and 9 antigens. We compare 20 computational approaches without task-specific training or fine-tuning. The analyses show that global structure-conditioned inverse folding yields the strongest and most consistent affinity rankings, whereas antibody-specific fine-tuning does not consistently improve performance. Across models, rankings are more useful for prioritizing likely affinity gains than for excluding weak variants. In a subsequent case study, multi-objective evaluation of model-generated F045-092 CDR-H3 variants for H1N1 binding revealed a trade-off between predicted binding potential and evolutionary sequence plausibility; all 21 selected candidates produced H1-reactive ELISA signals. AbBiBench provides an open framework for developing and evaluating function-aware antibody design methods.

q-bio.QM↗
arXiv2026

Maternal Anxiety During Pregnancy and Offspring Information Processing Across Development: A Theory-Guided Predictive-Processing Reappraisal of Two Longitudinal Cohorts

Prenatal maternal distress has been associated with differences in offspring brain and behavioural development, but is rarely framed computationally. Predictive-processing (PP) accounts propose that perception, cognition, and action arise from interactions between input and internally generated predictions, partly governed by precision. We reappraised eleven publications from two non-clinical prospective cohorts, selected using prespecified criteria, to assess whether they form a coherent PP-compatible developmental pattern. Eight Leuven publications spanned adolescence to adulthood. Externally cued response inhibition showed no detectable group differences in modestly powered samples, whereas differences emerged when task goals, stimulus relevance, expected outcomes, and response policies had to be generated or maintained endogenously. At age 28, higher prenatal anxiety was associated with lower radial and mean diffusivity in the left superior posterior corona radiata, lower Vocabulary scores, weaker medial prefrontal-left inferior frontal and prefrontal-left supramarginal gyrus connectivity, and greater unpleasantness of induced dyspnoea. In three Tilburg ERP publications (infancy and early childhood), higher prenatal anxiety was associated with increased processing of repeated standard tones, fearful relative to happy vocalizations, and neutral rather than pleasant or unpleasant pictures. Maternal mindfulness showed the opposite association for the repeated standard. Across cohorts, findings were compatible with differences in how relevance and processing resources are allocated to predictable, threat-related, ambiguous, contextual, feedback-related, and bodily information. Because the publications did not estimate precision, prediction errors, or priors, this is a theory-guided, post-hoc reappraisal that generates discriminating hypotheses rather than a formal computational test.

q-bio.QM↗