> ## Documentation Index
> Fetch the complete documentation index at: https://docs.revilico.bio/llms.txt
> Use this file to discover all available pages before exploring further.
# Planning Retrosynthesis
> Plan Synthetic Routes and Pathways for Lead Compounds
## **The Problem You Are Trying to Solve**
*“I have a set of lead compounds, and I want to plan practical synthetic routes and pathways so I can prioritize what to make next.”*
At the lead optimization stage, the best molecule on paper is only valuable if it is buildable. Synthetic planning is often constrained by:
* Limited starting material availability
* Route length, yield, and step complexity
* Risky transformations or fragile intermediates
* Cost, cycle time, and scalability considerations
This workflow helps you move from candidate SMILES → actionable synthetic routes, with clear prioritization signals across your lead set.
## **Solution**
This workflow uses Revilico’s **Retrosynthesis engine** as the core route-planning layer, supported by optional feasibility and risk checks from other chemistry engines. The primary synthetic planning chain is: Lead Set Preparation → Retrosynthesis Route Generation → Route Ranking & Comparison → Starting Material & Risk Review → Export + Iterate.
Binding, property, and AI engines can be integrated to ensure you prioritize the compounds that are both valuable and makeable.
## **What Data Do I Need to Provide?**
Required
* Lead structures as SMILES (CSV upload or manual input)
Recommended
* Optional compound identifiers (name / series / project tag)
* Any “hard constraints” from your chem team (must avoid certain reagents, protect certain groups, limit step count, etc.)
Optional
* Known preferred intermediates or supplier catalogs (if your org has them)
* Target number of steps / cost ceilings / timeline constraints
## **Workflow**
1. **Prepare and Organize Your Lead Set**
Start by ensuring your lead list is clean and trackable. On Revilico, users typically:
* Upload a CSV of lead SMILES (plus optional IDs/names)
* Confirm structures are valid and standardized
* Group compounds into series (if relevant) so you can compare routes across analog families
This will give you a versioned lead set ready for route planning.
2. **Generate Retrosynthetic Pathways**
Use **Retrosynthesis** to propose diverse synthetic routes for each lead. This engine will:
* Expand multiple disconnection strategies per molecule
* Produce stepwise pathways with intermediates and reaction class labels
* Rank routes by a route-quality score (route plausibility, efficiency, starting material reasonableness)
This step answers:
* *How would I make this?*
* *How many routes exist?*
* *Which ones look most realistic?*
This will produce a ranked set of retrosynthetic pathways per lead.
3. **Compare and Prioritize Routes Across Leads**
Now shift from route generation to decision-making. Users typically review:
* Step count (shorter is usually faster and lower risk)
* Route diversity (multiple independent options reduces project fragility)
* Intermediate complexity (risk of bottlenecks)
* Starting material practicality (availability and cost proxies)
* Convergence opportunities (shared intermediates across a series)
* At this point, all of the data can be sent to your chemistry team for utilizing these newly generated hypotheses as a baseline for getting these molecules synthesized
This step is where you decide which molecules are:
* Ready to make now
* Worth minor redesign to reduce synthesis complexity
* Not currently practical relative to alternatives
This will give you a prioritized list of leads based on synthetic feasibility and route quality.
4. **Sanity-Check Molecular Feasibility and Stability (Optional)**
For routes that look promising but uncertain, validate that candidates and intermediates are physically reasonable.\
Optional supporting engines:
* **Geometry Minimization and Thermochemistry** to sanity-check stable geometries and identify strained or unstable candidates
* **Molecular Orbital Analysis (HOMO–LUMO)** to flag potentially reactive or unstable electronic profiles (helpful for identifying “looks good, but might be chemically problematic” cases)
* **Conformer Search** to highlight extreme flexibility or conformational strain that could complicate synthesis or isolation
This will give risk flags and confidence boosts on route feasibility.
5. **Export Routes and Create a Make-List**
Once routes are selected, generate outputs that enable execution:
* Route summaries per compound
* Stepwise reaction outlines and intermediates
* A consolidated “Make Next” list for your synthesis team
This step is designed to reduce handoff friction from computational planning → wet lab execution with a synthesis-ready route package and execution shortlist.
**Now what?**
* With the data on hand for planning synthesis, your chemists can now move forward with getting the molecules synthesized.
* If you are using this engine as a secondary screen to any other engine, you can utilize the results as a sanity check of the compounds to ensure that the compounds made with generative chemistry are feasible to move forward with.
## **Integration with Other Engines (Optional)**
Synthetic planning rarely happens in isolation. Revilico supports tight integration with:
* ADMET-AI + Solubility to avoid planning routes for compounds likely to fail developability
* Docking / MD / FEP to ensure synthesis effort is directed toward leads with strong on-target justification
* Generative Chemistry to redesign hard-to-make leads into more synthesizable analogs while preserving activity motifs
## **Why Revilico?**
Revilico makes synthetic planning actionable by combining:
* A dedicated retrosynthesis engine for route generation and ranking
* Optional physics/QM checks for stability and feasibility confidence
* AI-assisted interpretation to speed prioritization and iteration
* Integrations with design, screening, and optimization workflows so you can plan synthesis for the *right* compounds, not just the most interesting ones on paper.