Reaction condition intelligence · starting with Suzuki–Miyaura

Coupling conditions ranked by who has successfully reproduced them.

You made the substrate. ReproRank shows you which protocols independent organisations have already made work on chemistry like it, with the evidence behind each one.

live search · October 2026
Query · 4-bromoanisole + phenylboronic acid
1
Pd(dppf)Cl₂ · K₂CO₃ · dioxane/H₂O · 90 °C
6 organisations · yield 100%REPRODUCED
2
Pd/PAd₃ · K₃PO₄ · THF/toluene · 25 °C
1 organisation · yield 81%SINGLE SOURCE
3
Pd-PEPPSI/NHC · K₂CO₃ · EtOH · 25 °C
1 organisation · close analogsSINGLE SOURCE
Every card links to the source patent and draws the precedent's structures.
Structures never storednot even as a hash
Hosted in the EUHelsinki, Finland
Built by a medicinal chemistfor medicinal chemists
The challenge

Search engines give you hits.
You need a decision.

Forty published condition sets for a substrate like yours, and forty milligrams of material six steps in. Which one do you run first?

The challenge

  • Hundreds of unranked hits.A wall of examples per substrate class, with no order of trust.
  • Conditions rarely travel as printed.What worked on one team's substrate often fails on the next.
  • A yield is not robustness.A single 92% says nothing about whether it transfers to your molecule.
  • Hours of triage per coupling.Time spent reading, not making compounds.

How ReproRank solves it

  • The best-supported protocol first.Plus a backup or two, ready to set up.
  • Evidence you can check.Each recommendation shows how many independent organisations used it, and its source.
  • Failed results shown as failed.Poor results are flagged, never recommended.
  • Seconds per query.With the source patent behind every row.
What makes the ranking different

One precedent is an anecdote.
Several independent ones are evidence.

When unrelated teams have made the same protocol work on similar chemistry, it has been tested by many hands.
ReproRank brings that evidence to you.

Reproduced, not just published

Protocols that other teams have already made work on chemistry like yours come first.

Complete procedures only

Catalyst, ligand, base, solvent and temperature, so what you read is what you can set up.

Traced to the patent

Every precedent is drawn as structures and linked to its source document.

Honest about gaps

Thin precedent is said out loud. Distant matches are labelled as leads, never dressed up as consensus.

How to use it

From your substrate to a bench-ready protocol.

Three steps. No keyword guessing, no list of 200 hits to sort through.

01
QUERY

Enter your coupling

Paste your electrophile and boron partner, or the product. ReproRank looks for precedent on chemistry like yours.

02
RANK

Rank the evidence

Complete procedures, best-supported first, with the evidence behind each one.

03
RUN

Go to the bench

Leave with the best-supported protocol to try first and a backup or two. Set it up and go.

Who it's for

Teams that need the right conditions the first time.

Medicinal chemistry teams

Late-stage couplings on scarce intermediates, where a failed first attempt costs days and material you cannot remake quickly.

CROs

Many client chemistries, tight timelines. Start each campaign from precedent that has already moved between labs.

Biotech startups

Small chemistry teams without an information department, making literature-grade decisions in seconds.

The dataset

Curated by a chemist.

The first dataset covers Suzuki–Miyaura couplings from pharmaceutical patents, checked by a medicinal chemist.

19,547
Curated reactions
2,161
Pharma patents
2003–25
Coverage window
77%
Structure-searchable
REPRODUCED

The strongest evidence: many independent organisations ran this protocol on substrates like yours.

CONFIRMED

Repeated by other organisations, outside its first source.

SINGLE SOURCE

One organisation. Real precedent, not yet confirmed.

DISTANT PRECEDENT

No close match. Shown as a lead, never as consensus.

Matthieu Desroses
Built at the bench
“I have spent over twenty-five years in medicinal chemistry, and a painful share of it repeating literature conditions that worked on paper and refused to work at the bench. ReproRank answers one question: of everything published, what has been reproduced, and is most likely to work in your hands?”
Matthieu Desroses
Founder · Medicinal chemist · Stockholm, Sweden
Pilot

Try ReproRank,start with couplings you know well.

The pilot is free.You get a personal link and can run as many searches as you like for 30 days.Published or patented chemistry is ideal; nothing confidential is needed.

pilot@reprorank.comStructures and names you enter are never stored, not even as a hash.Hosted in the EU.Already a pilot? Open the personal link you received again to get back in.