Research evidence index

Keep each claim attached to its method, figure, source, and limit

This index routes readers to the evidence-bearing chapters of FROGENT and every product site. Scientific evidence, manuscript positioning, and company-level portfolio language remain separate layers.

01 / FROGENT manuscript evidence

Architecture, eight benchmark families, and three end-to-end case studies

The supplied manuscript compares FROGENT with six progressively capable baselines across task-specific benchmarks and presents three computational design case studies.

FROGENT benchmark atlas comparing the system with six baselines across eight drug discovery task families
Benchmark atlas / task-specific definitions and scales remain attached to the manuscript figure.
FROGENT cardiomegaly and congestive heart failure case-study workflow from retrieval to synthesis planning
End-to-end case study / a computational trail from disease question through synthesis planning.
Case 01

Cardiomegaly and congestive heart failure

Target retrieval, structure acquisition, generation, screening, lead optimization, and retrosynthesis.

Case 02

GLP-1 receptor peptide optimization

A design-evaluate-refine loop for peptide candidates under a consistent in silico protocol.

Case 03

Carbonic anhydrase inhibitor optimization

Structure analysis, analogue generation, evaluation, and synthesis-path repair using retrieved evidence.

Evidence boundary. These entries summarize the supplied manuscript. Computational evaluation does not establish experimental efficacy, safety, affinity, or clinical performance.

02 / Product evidence

Direct routes to results, limitations, sources, and downloads

01

Scientific Design Engine

TrioMol2

All-target molecular generation

Scope. All-target campaign, AML1–ETO9a pocket study, and OpenBind diagnostics.

Evidence focus. Coverage ledger, ten-round signals, contact fingerprint, curated chemistry, backend stability, state aggregation, retrieval gap, and limitations.

02

Scientific Design Engine

TrioPROTAC

Structure-aware degrader design

Scope. Fixed-anchor degrader design and archived computational studies.

Evidence focus. B5/C0 anchors, linker geometry, constrained search, restricted docking, 100-search acceptance record, weighting variants, replay validity, and source boundaries.

03

Scientific Design Engine

TrioPep

Structure-feedback peptide design for XOD

Scope. Structure-feedback peptide design for xanthine oxidase (XOD).

Evidence focus. Reward decomposition, fixed-budget design, candidate curation, interface evidence, reported activity context, and three explicit evidence limits.

04

Scientific Design Engine

TrioIRES

Structure-constrained RNA design

Scope. Structure-constrained internal ribosome entry site (IRES) prioritization.

Evidence focus. CrPV, TSV, and IAPV fixed positions, secondary-structure maps, resolved-chain coverage, candidate comparisons, source links, and limitations.

05

Scientific Design Engine

TrioDNA

Cell-type-aware regulatory DNA design

Scope. Cell-type-aware regulatory DNA design across enhancer and yeast-promoter evaluations.

Evidence focus. Reward difference, transcription factor binding site motif correlation, generation-budget sensitivity, diversity comparison, provenance, tables, and CSV download.

06

Scientific Agent

Pio

Evidence-aware peptide optimization

Scope. Evidence-aware peptide optimization and review contracts.

Evidence focus. Runtime components, ten-stage workflow, executable capabilities, tools, governed memory, trace ledger, three reviewers, task contract, evidence, and test plans.