Biology · Biological Frontiers & Open Questions

Open questions, new evidence and responsible research directions

A research question remains open when a relevant mechanism or inference is not yet settled. That does not mean every explanation is equally supported or that nothing useful is known. A clear starting point identifies the claim, the alternatives, what each predicts and which observation could make the difference.

Earth, Origins & the History of Life introduces the difference between biological history and processes possible today. Development, Differentiation & Stem Cells introduces cell states and organismal organisation. Genomics, Omics & Bioinformatics distinguishes molecular records, identity, proportions and validation. Systems & Quantitative Biology supplies state, observation and model-comparison reasoning. Each lesson also explains its local terminology; the original examples require only fractions, means and signed differences. Follow the concrete cases before returning to the broader research question.

Back to the Biology map

By the end, you can

Turn an open biological question into explicit alternatives and evidence needs, and connect responsible research choices with the strength and limits of the resulting claim.

Make the uncertainty specific

An open question should identify its scale and target: a molecular process, a cell state, an organismal function, a community interaction or a historical event. The form of evidence that is useful depends on that target. Measuring more of an adjacent quantity can be valuable while leaving the actual question unanswered.

Alternative explanations make an evidence need concrete. In the earlier cell example, the question of regional arrangement remained open after the bulk mean was known. In the microbiome example, an association fit two causal routes. In the origin example, partial demonstrations left an integration or historical question. Each uncertainty points to a different missing link.

The claim ledger distinguishes local support, compatibility, contradiction and an unmeasured prediction. These words concern the relationship between a stated claim and its included evidence. They are not permanent ranks for entire disciplines or a numerical certainty scale. A useful local result can still leave a broader mechanism unsettled.

Different evidence relations deserve different words

Good research makes the proposed claim, its alternatives and the evidence that could change it easy to inspect.

Read the diagram in words

Claims 1 and 2 both match a positive record, but only claim 1 includes a specified local discriminating test; claim 2 shares its record with an alternative. Claim 3 predicts absence while the supplied valid record shows presence. Claim 4 has no observed test result.

Relation between this claim and its evidence
Anonymous claim and its test context	Assigned prediction	Included observation	Prediction of the included alternative where supplied	Relation between this claim and its evidence
Claim with a matching included test	true	true	false	Locally supported under the stated test
Claim sharing a record with another candidate	true	true	true	Compatible; the record does not distinguish the candidates
Claim whose stated prediction disagrees	false	true	—	Contradicted under the supplied measurement and assumptions
Claim with an unmeasured prediction	true	?	—	Not tested by the included record

Original qualitative records for four anonymous claims, not a universal evidence ranking or probabilities. Each status belongs to its specified observation/model pair. Supported means a stated local prediction was observed under the included conditions; it does not imply a unique mechanism or final truth. Compatibility has no discriminating observation, contradiction assumes a valid measurement and the tested prediction, and unmeasured remains unknown.

Explain what would change the claim

A discriminating prediction differs among the relevant candidates. A new record tests that difference only when the observation and assumptions are appropriate. An unexpected result may challenge a candidate, an observation rule or an assumption shared by several candidates. Revising the explanation requires identifying which part was tested.

Reproducible reasoning includes access to the claim, the relevant evidence and enough description to understand the comparison. Open records, code and community scrutiny can support that process when appropriate. They do not make an unsupported claim true, and disclosure choices still need to consider privacy and the people or environments affected.

The reasoning graph returns from a revised interpretation to the question. Research can narrow a claim, reveal another alternative or identify that the available observation cannot decide the issue. An inconclusive result still carries information about the limits of the included test; it need not be relabelled as positive evidence for a preferred explanation.

A new record can change a claim without changing the original question

Responsible research connects evidential strength with the claims and actions it can justify, including who is affected and what remains uncertain.

Read the diagram in words

The question motivates alternatives. Their differing predictions guide a relevant record. The record can support,reject or leave a claim unresolved under its conditions. The revised interpretation can return to the question with a narrower claim or a new evidence need.

Meaning of the information relationship
Role in a research question	Role in a research question	Meaning of the information relationship
Bounded biological question	Explicit alternative explanations	Identify explanations for this question
Explicit alternative explanations	Prediction that differs between candidates	Compare what the candidates predict
Prediction that differs between candidates	New record with observation conditions	Evaluate a relevant observation
New record with observation conditions	Revised claim and remaining uncertainty	Update the stated evidence relation
Revised claim and remaining uncertainty	Bounded biological question	Refine the question or plan further evidence

Original information graph representing scientific reasoning, not material flow or a causal biological pathway. Nodes are editorial/scientific roles. An edge indicates a reasoning relationship, not a guarantee of an answer or a universal linear research method. An inconclusive record may lead to revising the question,method or candidates rather than selecting a winner.

An evidence claim and an action have different requirements

Research directions are shaped by scientific importance, tractability, resources and the possible benefits and burdens of the work. These considerations are related but distinct. A promising hypothesis can justify investigating a question without already justifying a public claim of effectiveness or broad application.

Responsible decisions ask who supplies data or materials, who bears risks, who can use the resulting knowledge and whose questions are represented. The answer is specific to the work. An open resource can widen participation, while missing representation or unequal access can limit how well its findings serve the intended population.

A useful entry point into frontier biology is therefore a bounded question that can be connected to existing concepts and inspected evidence. The next step should preserve what is already known, identify what remains uncertain and explain how new evidence could alter the interpretation. Scientific openness includes being explicit about those limits.

Worked example

Choose the evidence relation that the record supports

A supplied positive observation agrees with claim 1 under its specified local test. Claim 2 also predicts a positive record, but an alternative agrees as well. Calling claim 2 compatible preserves that unresolved discrimination.

Claim 3 predicts absence while the supplied valid observation shows presence. Under the stated assumptions that is a contradiction. Claim 4 predicts presence but has no observed result; it remains untested rather than false.

A revised report can state the local outcome, the conditions and the remaining candidates. It can then identify the next evidence need without turning the current ledger into a final verdict or a universal certainty score.

Check your understanding

Does an open question mean that all its explanations are equally supported?
Show reasoning
No. Some candidates can match discriminating evidence, some can be merely compatible, some can be contradicted and others untested. The status needs to be stated for the actual claim and record.
Why is an unmeasured prediction not a negative result?
Show reasoning
A negative result would require a relevant observation under appropriate conditions. If no record was supplied, the prediction has not been tested by this evidence and its observation stays unknown.
Does making a dataset open establish that its biological claim is correct or universally applicable?
Show reasoning
No. Openness can allow inspection and reuse, while correctness and applicability still depend on the observation, analysis, alternatives and represented domain. Privacy, participation and access also remain distinct considerations.

What to carry forward

A strong research starting point connects a precise question, explicit alternatives, relevant evidence and an honest account of what could change the claim.

References

Resources supporting this lesson’s explanations and examples.