Platform
Computer aided drug design, narrowed to one question
Most computer aided drug design software is general purpose: a target, a pocket, a docking score. Antibacterial is narrow on purpose. Everything here exists to answer one question well, which of these compounds is worth a plate against which strain.
Capabilities
Six things, done properly
Strain panels
ESKAPE, Gram-negative and Gram-positive sets out of the box, custom panels from Discovery, and your own internal isolate library on Enterprise. The panel is the axis of everything else.
Predicted MIC bands
A doubling dilution band per strain in µg/mL, matching the resolution a broth microdilution plate actually has. Never a single decimal pretending to be a measurement.
Resistance risk flags
The mechanism the model expects, named: efflux and which pump, a carbapenemase family, porin loss, a target mutation, an acquired operon. A flag without a mechanism is not actionable.
Analog and literature lookup
The closest published analogs the model can genuinely name, with their reported activity, so you can see what the prediction is standing on.
Batch runs and ranking
Upload an SDF or CSV and get the series ranked by how much of the panel each compound holds, with the losses attributed to a mechanism.
Exports and API
CSV everywhere, SDF and a PDF report from Discovery, scheduled exports on Program. A REST API from Discovery so results land in your registry, not in a screenshot.
Strain panels
The axis of the whole product
A target based tool asks whether a compound binds. An antibacterial program has to ask whether it works on an organism that has an outer membrane, three efflux pumps and a plasmid. That is a different axis, and it is the one every read-out here is organized around.
Custom panels arrive on Discovery. If your collection has its own isolates with their own resistance profile, Enterprise loads them so the matrix matches your bench.
Why Gram-negatives are the hard caseESKAPE
Six pathogens behind most resistant hospital infections.
- S. aureus (MRSA)
- E. coli
- K. pneumoniae (CRE)
- A. baumannii
- P. aeruginosa
- E. faecium (VRE)
Gram-negative
Permeability and efflux decide almost everything here.
- E. coli
- K. pneumoniae (CRE)
- A. baumannii
- P. aeruginosa
- E. cloacae
- S. Typhimurium
Gram-positive
Target modification and acquired operons dominate.
- S. aureus (MRSA)
- S. aureus (MSSA)
- E. faecium (VRE)
- E. faecalis
- S. pneumoniae
- S. epidermidis
The category
Five kinds of computer aided drug design software and what each one answers
A program usually owns two or three of these. The mistake is buying a docking or PBPK seat to answer a question it was never built for, then finding out a year later that nobody knew which compounds were active on the organism in the first place.
Every product named here has its own page on this site with the license figures we could document from public US federal awards, so you can price the alternative before you choose.
Drug discovery software cost, comparedDocking and modeling suites
AnswersDoes this compound bind this target, and in what pose
NeedsA protein structure, a prepared pocket and a modeler to run it
QSAR and ADMET prediction
AnswersPhysicochemical, ADME and toxicity endpoints, or a model you train
NeedsYour own consistent assay data for any potency model
PBPK and PK modeling
AnswersWhat exposure a dose produces over time
NeedsIn vitro and PK inputs, plus an MIC as the denominator of every PK/PD target
Cheminformatics and workflow
AnswersRegistration, search, structure data and automated pipelines
NeedsAn informatics owner and data to organize
Antibacterial screening
This platform
AnswersWhich strains a compound inhibits, at what MIC band, and what will take it out
NeedsA structure, or an SDF of the whole series
Anatomy of a read-out
Five fields per strain, and what each one is for
| Field | Example | What it is for |
|---|---|---|
| Predicted MIC band | 0.5-2 µg/mL | The concentration range where growth is expected to stop, at plate resolution. |
| Call | I Intermediate | A susceptible, intermediate or resistant read at the usual clinical breakpoint for that species. ND when there is no basis. |
| Resistance risk | efflux, MexAB-OprM | The mechanism most likely to take the compound out on that organism. This is the line that kills or saves a scaffold. |
| Reason | One sentence | Why the call is what it is, in language you can paste into a program meeting. |
| Confidence | 3 of 4 | How much published support sits behind the row. Low confidence is shown, not hidden. |
The whole read-out carries one standing label: computational prediction from published literature, research use only. See the method for how each field is derived.
Getting data in and out
Nothing you put in is locked in
Endpoints, payloads and refusal codes are written up on the antibacterial screening API page.
- Input
- SMILES, InChI, compound name, SDF, CSV
- Batch size
- 50 on Lab, 1,000 on Discovery, 10,000 on Program
- Export
- CSV, SDF, PDF report, scheduled exports on Program
- API
- REST with an API key, from Discovery upwards
- Rate
- Plan level monthly screen allowance, no per call surcharge
Governance
The part your chemistry team will ask about
- Training
- Submitted structures are never used to train any model
- Sharing
- Never shared with other customers
- Deletion
- Deletable on request
- SSO
- SAML 2.0 and OIDC on Program and Enterprise
- Audit
- Full audit log on Program and Enterprise
- Deployment
- Private or VPC deployment on Enterprise
Seats and roles
Who on the team can spend the month, and who can only read it
A workspace holds 2 seats on Lab, 8 on Discovery and 25 on Program, and the monthly screen allowance belongs to the workspace rather than to a person. That is the detail procurement asks about, because a supervisor, a partner or a finance reviewer who only needs to read a matrix should not be able to spend a chemist's month. From Discovery upwards every invited seat carries one of three roles, and the viewer role exists exactly for that case.
The table is the permission list the application itself enforces, not a marketing summary of it. Seat counts and allowances per plan are on the pricing page, and the same roles govern key creation on the screening API.
| What the seat can do | Owner | Analyst | Viewer |
|---|---|---|---|
| Run a single screen | ✓ | ✓ | · |
| Run a batch | ✓ | ✓ | · |
| Build and edit strain panels | ✓ | ✓ | · |
| Load your own clinical isolates | ✓ | ✓ | · |
| Export results as CSV, SDF or PDF | ✓ | ✓ | ✓ |
| Schedule a recurring export | ✓ | ✓ | · |
| Read the audit log | ✓ | · | · |
| Invite and remove people | ✓ | · | · |
| Create and revoke API keys | ✓ | · | · |
| Set up single sign on | ✓ | · | · |
| Change the plan and the card | ✓ | · | · |
On Lab every seat behaves as an owner, because a two seat workspace does not need an internal permission boundary. Roles start where the team does.
Try it here
The product, not a picture of it
This is the same screen that runs on the homepage. Change the compound, change the panel, and read the matrix.
Run a screen
of 3 left
Strain panel
No account needed. 3 screens per session.
| Strain | MIC (µg/mL) | Call | Resistance risk | Conf. |
|---|---|---|---|---|
| S. aureus (MRSA) | 0.5->32 | R Resistant | target mutation, grlA and gyrA Fluoroquinolone resistance is widespread in methicillin resistant lineages. | |
| E. coli | <=0.015-0.06 | S Susceptible | efflux, AcrAB-TolC in resistant isolates Potent against wild type; qnr carriage and gyrA changes shift the band sharply. | |
| K. pneumoniae (CRE) | 0.5->64 | R Resistant | gyrA mutation with plasmid qnr Carbapenemase producing isolates almost always carry quinolone resistance too. | |
| A. baumannii | 8->64 | R Resistant | efflux, AdeABC Constitutive efflux plus target changes leave little room at achievable exposure. | |
| P. aeruginosa | 0.25-2 | I Intermediate | efflux, MexAB-OprM Borderline: active on many isolates, lost quickly once efflux is derepressed. | |
| E. faecium (VRE) | 4->32 | R Resistant | target mutation, parC Enterococci are intrinsically poor fluoroquinolone targets. |
Why
Ciprofloxacin is a well characterised fluoroquinolone, so the Gram-negative bands are strongly supported by published activity data. The deciding factor across this panel is not target affinity but exposure: efflux in P. aeruginosa and A. baumannii, and acquired target mutation everywhere resistance is already common. Against a modern ESKAPE panel it reads as a Gram-negative agent with two reliable losses.
Closest published analogs
- Levofloxacin Broadly similar Gram-negative bands, better Gram-positive coverage
- Delafloxacin Retains activity against many ciprofloxacin resistant staphylococci
| Strain | MIC (µg/mL) | Call | Resistance risk | Conf. |
|---|---|---|---|---|
Why
Closest published analogs
Want the rest of your series? Batch runs are on the paid tiers. screen left in this session. This session is used up.
Computational prediction from published literature. Research use only, not a lab measurement and not clinical guidance.
Related pages
- How it works The four step method in detail
- Use cases Biotech, CRO, academic lab, natural products
- Pricing Four tiers and the full feature matrix
- Minimum inhibitory concentration The number the read-out predicts
- Antimicrobial resistance Surveillance and the discovery gap
- ChEMBL data What public bioactivity data can and cannot answer
Bring a series, not just a compound
Batch upload ranks the whole set and attributes each loss to a mechanism. Start with one screen on the homepage, then create your account for the rest.
What you are agreeing to
- Your compounds stay yours
- Never used for model training
- Deletable on request
- No card needed to create your account