Skip to content
antibacterial .ai

Compound versus strain screening

Antibiotic discovery without the dead ends

Screen a compound against MRSA, CRE and the rest of the ESKAPE panel before you book plate time.

Run a screen

of 3 left

Strain panel

No account needed. 3 screens per session.

Sample output Ciprofloxacin Fluoroquinolone
Worked example, replaced when you run
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

Save this screen and run the rest of your series. 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.

01

Kill dead-end scaffolds at week two

A named resistance mechanism is a defensible reason to stop. Without one, a series gets carried for another three months on hope and half a dataset.

02

Know why it fails, not just that it fails

Every call comes with the mechanism the model expects, the closest published analogs, and a confidence level. Low confidence says low confidence.

03

One paste, six strains, ten seconds

A structure or a name in, a strain by read-out matrix out. The afternoon of ChEMBL querying happens once, in the model, not once per compound in your head.

How it works

Four steps between a structure and a decision

The method in detail
01

Paste a structure or a name

A SMILES string, an InChI, or just the compound name if it is a known agent. Batch upload takes an SDF or a CSV on every paid plan.

02

Pick the strain panel

ESKAPE, Gram-negative, Gram-positive, or a custom set of clinical isolates. The unit of work is compound times strain, not compound times target.

03

The model reads the published neighbourhood

Reported antibacterial activity for the compound and its closest structural analogs, plus what the surveillance literature says about resistance in each species.

04

You get a ranked read-out with confidence

Predicted MIC band, an S, I or R call, the expected resistance mechanism, and how much published support each row has. Export it or send it to the API.

Worked example, ranking a series

Four fluoroquinolones screened against the same panel, ordered by how much of it each one holds. This is what a batch run returns per compound.

Compound Panel held Best band
Delafloxacin 4 of 6 <=0.008-0.12
Moxifloxacin 3 of 6 0.015-0.06
Levofloxacin 3 of 6 0.015-0.12
Ciprofloxacin 2 of 6 <=0.015-0.06

Computational prediction, research use only

Where it fits

A screening layer, not another place to keep data

Discovery suites are built for target-based small-molecule work and informatics platforms are built to store what you already measured. Neither has a concept of a strain panel or a breakpoint. Public databases have the data and no opinion about it.

Capability Antibacterial General discovery suite Manual ChEMBL querying
Unit of work Compound times strain Compound times target Assay record
Predicted MIC band per strain Yes, with confidence No Only what was measured
Resistance mechanism named Yes, per strain No You infer it yourself
Closest published analogs Returned with the result Similarity search, no activity roll-up One query per analog
Time for one compound About ten seconds Hours of setup per project An afternoon
Onboarding A text field Install, licensing, training None, and no workflow either
Price anchor From $149 a month Enterprise licence Free

For heads of discovery

The expensive failure is the series you carried too long

Check the arithmetic with your own numbers

A confirmatory MIC panel at a CRO takes about a week of calendar time and a price only you know. Put yours in and compare it against the plan lines.

Sent out per month
Plan Per month In your panels

Your figures, not ours: we do not know your CRO's price list, and a screen does not replace the panel you finally run. It changes which compounds get one.

Kill at week two, not month four

A ranked list with named mechanisms gives a program lead something to argue with. Dead-end scaffolds usually survive on the absence of a reason to stop, not on the presence of a result.

Governance your chemistry team will accept

Submitted structures are never used to train anything, are not shared, and are deletable on request. For a discovery team, that is usually the whole conversation.

It is a prediction, and it says so

Every row carries a confidence level and the research-use-only label. A screen tells you what is worth measuring. The plate still tells you what is true.

See how teams use it

Enterprise signals

  • SSO SAML 2.0 and OIDC
  • Access Roles and permissions
  • Audit Full audit log
  • SLA 99.9% on Program and above
  • Deployment Private or VPC on Enterprise
  • Legal DPA on request, invoicing and PO
  • Residency Data residency options
  • Training Your structures are never used
Read the security page

Pricing

Priced against one confirmatory panel

Full feature matrix

Lab

$149 /mo

$124/mo billed yearly ($1,488)

250 screens a month against the standard panels, with CSV export.

Academic AMR lab, 1 to 2 chemists

Discovery

Most popular

$499 /mo

$416/mo billed yearly ($4,992)

Custom panels, batch runs of a full series, API access and roles.

Biotech antibacterial program

Program

$1,490 /mo

$1,241/mo billed yearly ($14,892)

Library-scale runs, your own isolates, SSO, audit log and an SLA.

Multi-program biotech or discovery CRO

Enterprise

Talk to sales

Custom terms, invoiced

Private deployment, your internal isolate library, a named contact.

Pharma, institute or consortium

Questions

The honest answers first

A prediction. Antibacterial estimates how a compound is likely to behave against a strain, based on published antibacterial activity data and the structural neighbourhood of the compound. It is not a measurement, it does not replace a plate, and every row of every result carries a confidence level and the standing label "computational prediction, research use only". A screen tells you what is worth measuring. The plate still tells you what is true.

A ChEMBL query returns records. A screen returns a decision. Antibacterial rolls activity up per strain rather than per assay, names the resistance mechanism it expects and why, ranks the compounds in your series against each other, and does it in about ten seconds instead of an afternoon. ChEMBL is one of the sources underneath, and the honest answer is that if you only ever screen one compound a month, the manual query is free.

No. Antibacterial is a research tool for discovery teams. It is not a diagnostic, not an antibiogram, and not clinical guidance. Nothing it outputs should reach a patient decision, directly or indirectly. This is stated in the product, in the exports and in the terms.

No. Submitted structures are never used to train or fine-tune any model, they are not shared with other customers, and they are deletable on request. This is a written commitment in the terms, not a preference setting.

The ESKAPE panel (S. aureus including MRSA, E. coli, K. pneumoniae including CRE, A. baumannii, P. aeruginosa, E. faecium including VRE), plus standard Gram-negative and Gram-positive clinical isolate sets. Custom panels are available on Discovery and above, and Enterprise can load your own internal isolate library.

All questions

Get started

Which twelve of your four hundred go on the plate?

Run a screen on the compound you are arguing about this week. If the read-out is useful, create an account and bring the rest of the series.

No card required

Screen your own compound.
No card required.