引導式測驗 · 中等 等級
AI逆合成規劃測驗
了解逆合成搜尋、反應模板、路線排名、評估限制以及化學和實驗審查的需要。
問題 1 的 8
What does retrosynthesis planning predict from a target molecule?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- What does retrosynthesis planning predict from a target molecule?
- A predicted product from a forward reaction
- A ranking of solvents and conditions for an already specified reaction
- Possible precursor molecules and reaction steps leading to the target
- An experimentally measured yield for each proposed step
- How do template-based systems generate reaction suggestions?
- Rank possible routes using a search policy without applying chemistry transformations
- Apply learned or curated reaction patterns to molecular structures
- Generate three-dimensional conformers of the target molecule
- Estimate target toxicity from molecular features
- Why can a short predicted route still be impractical?
- Step count determines experimental success exactly
- All short routes use common reagents
- Route length is the only important criterion
- It may require unavailable, hazardous, unstable or selective reagents
- What does top-k one-step accuracy measure?
- Whether a reference precursor appears among the top k suggestions
- Whether every route contains the same number of steps
- Whether a predicted route was included in the training set
- Whether the target structure parses as a valid molecular string
- Why can reaction-database splits by random rows overstate generalization?
- Random splitting removes all chemical similarity
- Related reactions are always isolated within a single partition
- Related reactions or scaffolds may appear in both training and test sets
- Random splitting guarantees greater route diversity
- What can forward reaction prediction add to a proposed retrosynthesis route?
- Proof that a lab reaction succeeds
- A model-based consistency check for predicted steps
- Guaranteed yield and purity
- A replacement for reagent safety review
- What should a chemist review before attempting a suggested route?
- Only the route score
- Only the shortest path
- Only whether the target SMILES parses
- Reagents, conditions, stereochemistry, safety and starting-material availability
- Which statement about AI retrosynthesis is accurate?
- It proposes hypotheses that require chemical and experimental validation
- It guarantees every reaction step works
- It replaces synthetic chemists
- It predicts only the final molecule identity