BioinvestGPT User Guide — Clinically Derisk BIC Drugs (ENHANCEMENT)

Optimize drug candidates across 14 modalities with predicted developability profiles before committing to a Candidate Drug Profile.

Part A — Assess Drug Developability

Use the ENHANCEMENT workflow to optimize drug candidates across 14 modalities with predicted developability profiles before committing to a Candidate Drug Profile (CDP). Part A happens in the Generator's drug-centric view: you pick a seed drug and declare which developability parameters you want to move. Part B (below) carries those choices into the Validator, where each goal becomes both a design objective and a GO/NO-GO gate.

Open the drug-centric view

From the Hub, click Start on the Drugs card. The Generator opens to the drugs panel. Click any drug node to view its details: SMILES structure, mechanism of action, known targets, and side effects.

90 blockbuster drugs are enriched in the graph with revenue data and vulnerability scores, so you can start from a clinically proven scaffold rather than a blank canvas.

Toggle developability parameters

In the drug Details Modal, toggle the 6 developability parameters you want to optimize. Each one has a fixed optimization direction:

Parameter	Direction	What It Means
Oral Bioavailability	Up-Regulate	Improve oral absorption
CNS Penetrability	Up-Regulate	Improve brain penetration (for neuro indications)
Half-Life	Up-Regulate	Extend duration of action
Immunogenicity	Up-Regulate	Reduce anti-drug antibody risk (biologics)
Effect Sizes	Up-Regulate	Improve predicted clinical efficacy
Adverse Effects	Down-Regulate	Reduce predicted side effects

Click Confirm All Enhancements to send your seed drug and selected goals to the Validator's ENHANCEMENT workflow.

Tip: Pick only the parameters you genuinely need to move. Every goal you select becomes a GO/NO-GO gate the final BVCT verdict must clear, so over-selecting makes a GO harder to earn.

Where this leads: Confirming enhancements opens the Validator with BVCT Mission set to ENHANCEMENT and your seed drug pre-filled. Continue in Part B.

Part B — Seed Drug Design

The Validator opens with BVCT Mission = ENHANCEMENT and your seed drug pre-filled from the Generator. This is where you turn a scaffold plus a list of developability goals into a fully specified design brief.

Enter the Seed Drug Design

Your seed drug design is pre-filled from the Generator. Review and complete these key fields:

Modality — Auto-detected from your input (small molecule, peptide, mAb, ADC, oligonucleotide, etc.)
Drug Design — SMILES string, amino acid sequence, or HELM notation
Target Protein — Primary target (e.g., EGFR), auto-detected from seed drug structure
Developability Goals — Checkboxes for the 6 parameters
Reference Drug — FDA-approved benchmark to compare against
TPP Route & Frequency — Oral, SC, IV combined with once daily, weekly, monthly
Current PK Limitations — What to improve (free text)
Known SAR Failures — Modifications to avoid (free text)

Required gate — at least one developability goal: ENHANCEMENT missions cannot proceed without ≥1 Developability Goal selected. The Predefine BVCT Protocols button stays disabled until at least one goal is checked. Each selected goal becomes two things at once: a differential design objective for variant generation, and a GO/NO-GO gate the BVCT verdict must clear.

If you arrived from Part A with goals already toggled, they carry through here and the button is enabled. If you started directly in the Validator, you must check at least one goal before continuing.

Modality-Specific Design Levers

Between the seed drug input and the developability goals, BVCT renders a Modality-Specific Design Levers panel — one of 14 panels purpose-built for the active modality. These levers constrain how variants are generated so the candidates stay inside your chemistry, manufacturability, and safety envelope.

Configure modality-specific design levers

The lever panel adapts to whichever modality is active. Examples:

Small molecule — target potency (IC50/Ki/Kd), hERG tolerance, CYP soft-spot aversion, Lipinski rule-of-5, MPO score floor, solubility, permeability, stereochemistry, genotox red-flag avoidance, synthetic accessibility cap, salt/solid-form preferences.
Antibody (mAb) — binding affinity, isotype, FcRn engineering, glycoengineering.
ADC — linker chemistry (cleavable / non-cleavable), payload selection, DAR target.
PROTAC / Degrader — E3 ligase preference (CRBN / VHL / IAP / DCAF15 / KEAP1), DC50 target, ternary complex stability, hook-effect tolerance.
Cell therapy (CAR-T) — CAR construct (hinge, costim, transduction efficiency), expansion, dose.
mRNA — cap analogue, codon optimization, polyA length, LNP composition.
Gene therapy — vector (AAV / lentiviral), capsid serotype, transgene, dose.

Lever values are persisted backend-primary — encrypted at rest with AES-256-GCM in the enhancement_briefs.design_levers_encrypted column. They populate the PDF appendix in the generated protocol so reviewers see the full Step-1 spec. Switching modalities preserves your prior modality's lever values per-modality, so you can compare designs across modalities without losing inputs.

Tip: Use the "Load preset" dropdown at the top-right of the lever panel to start from a curated lever profile (e.g., "GLP-1 BIC", "PROTAC default") and adjust from there.

Choosing your modality

Choose your modality

The platform supports 14 drug modalities, each with a purpose-built molecular editor:

Small Molecule (Ketcher + RDKit)
Peptide (bead-on-string editor)
Monoclonal Antibody (CDR editor)
Antibody-Drug Conjugate (antibody + linker/payload)
Oligonucleotide (backbone/sugar editor)
Bispecific Antibody, Nanobody, Cell Therapy, Gene Therapy, RNA Therapy, Vaccine, Radiopharmaceutical, Protein Degrader, Other

The modality you pick determines which lever panel and which molecular editor appear. For full editor walkthroughs, see the Modality Editor Reference chapter.

Variant Generation

Once your seed drug, developability goals, and design levers are set, BVCT generates a small slate of differential drug design variants. Each one explores a distinct trajectory through the design objectives you declared — not random analogs, but candidates that trade off your selected goals in different ways.

Review variant strata

BVCT generates 3-5 differential drug design variants spanning your design-lever space and selected developability goals — each variant explores a distinct trajectory through the differential design objectives you set. Each variant card shows: name, 2D structure thumbnail (RDKit), category, and annotations. Select the variants you want to advance to protocol generation.

Tip: Advance two or three contrasting variants rather than just the single best-looking one — the Comparison protocol (next section) is most useful when it has distinct candidates to place on the Pareto front.

Three-Audience Protocols

The Enhancement Protocol Panel generates protocols for three audiences simultaneously, plus a cross-variant comparison. Each audience gets the depth it needs — an executive does not read the SAR map, and a medicinal chemist does not need the cost bracket framed for a board.

Generate three-audience protocols

The panel produces four tabs:

Executive — GO/NO-GO badge, patent score, risk rating, cost bracket, time-to-IND, competitive advantage
Pharmacometrics — PK profile, dose-response, DDI risk, formulation strategy, population PK
MedChem — SAR map, synthetic route, selectivity, metabolic soft spots, stability, IND-enabling assays
Comparison — cross-variant heatmap, Pareto analysis, top 3 recommendations, failure mode analysis

Mission-specific Section 2 of the downloaded protocol PDF auto-populates from your workflow — drug modality, target protein, developability goals, reference drug, PK targets, and (for ADCs) antibody / linker / payload / DAR. No manual field entry needed.

Reading the verdict

Tab	Audience	Use it to
Executive	Program lead / board	Make the GO/NO-GO call and frame competitive positioning
Pharmacometrics	Clinical pharmacology	Sanity-check dose, exposure, and DDI assumptions
MedChem	Chemistry / DMPK	Plan the synthetic route and IND-enabling assays
Comparison	Whole team	Pick the winning variant off the Pareto front

A variant earns a GO only when it clears every developability goal you set in Part B. Because each goal is a gate, the Comparison tab's Pareto analysis is the fastest way to see which candidate clears the most gates with the fewest trade-offs. BVCT outputs are model-based decision-support analyses, not investment advice. Prospective track record: data.bioinvestgpt.com.

BioinvestGPT BVCT Platform User Guide — Clinically Derisk BIC Drugs (ENHANCEMENT). BVCT outputs are model-based decision-support analyses, not investment advice. Prospective track record: data.bioinvestgpt.com.