eduKateSG · Why Science?
Use precise genomic language so discovery, consent, privacy and medicine decisions stay connected
Read FDA’s final April 2008 ICH E15 guidance through genomic biomarkers, pharmacogenomics, pharmacogenetics, DNA and RNA characteristics, identified, coded, anonymised and anonymous samples, traceability and research design.
Genomic research can connect a DNA variant or RNA pattern with how a medicine is absorbed, how well it works or whether it causes harm. Shared definitions matter because a single word can change a laboratory plan, a consent conversation, a privacy safeguard and a regulatory interpretation.
FDA’s final April 2008 ICH E15 guidance harmonises definitions for genomic biomarkers, pharmacogenomics, pharmacogenetics and four broad categories of genomic data and sample coding. It helps global teams describe the same scientific and traceability concepts consistently.
Did you know? E15 distinguishes anonymised samples from anonymous samples. An anonymised sample once had a code link that was later deleted; an anonymous sample was never linked to personal identifiers and never had a coding key.
For students, E15 is a joyful meeting point of biology, data governance and ethics. The coding choice changes what future science and participant choices remain possible.
Chapter index
Chapters 1–9: April 2008 final guidance, genomic biomarker, DNA and RNA characteristics, PGx and PGt
- 1. E15 creates a harmonised vocabulary
- 2. The April 2008 FDA guidance is final
- 3. A genomic biomarker is a measurable DNA or RNA characteristic
- 4. DNA characteristics take many forms
- 5. RNA characteristics add dynamic information
- 6. Genomic biomarkers are not restricted to humans
- 7. Proteins and metabolites fall outside the E15 genomic definition
- 8. Pharmacogenomics connects DNA and RNA variation with drug response
- 9. Pharmacogenetics is a subset of pharmacogenomics
Chapters 10–17: identified, single-coded, double-coded, anonymised and anonymous samples, traceability, privacy and study design
- 10. The medicine concept includes biological products and vaccines
- 11. Identified samples are directly traceable
- 12. Single-coded samples use one key
- 13. Double-coded samples add another barrier
- 14. Anonymised samples lose the code link
- 15. Anonymous samples never had a coding key
- 16. Coding choice shapes consent and scientific options
- 17. A complete E15 receipt connects terminology to governance
The guidance defines key terms so regulators, ethics committees, sponsors, laboratories and researchers can interpret genomic work consistently across regions. Shared language reduces confusion in protocols, reports and submissions.
The clarity of e15 creates a harmonised vocabulary begins with naming the biological measurement and the data relationship precisely. The guidance defines key terms so regulators, ethics committees, sponsors, laboratories and researchers can interpret genomic work consistently across regions. Shared language reduces confusion in protocols, reports and submissions. Keep this boundary visible: E15 defines terms; it does not establish every validation, qualification or acceptance requirement.
Try this
Classify a biomarker study for e15 creates a harmonised vocabulary. Create a glossary and mark which questions sit outside the guidance’s scope. Separate analyte, response and intended use. Keep this limit: E15 defines terms; it does not establish every validation, qualification or acceptance requirement.
A strong science note separates observation from interpretation. Create a glossary and mark which questions sit outside the guidance’s scope. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Create a glossary and mark which questions sit outside the guidance’s scope. Keep this boundary visible: E15 defines terms; it does not establish every validation, qualification or acceptance requirement.
FDA lists E15 as final guidance dated April 2008 under docket FDA-2008-D-0199, issued by CDER and CBER. FDA states that consistent terminology supports integration of pharmacogenomics and pharmacogenetics into global drug development and approval.
For the april 2008 fda guidance is final, added privacy can remove future scientific or participant options. FDA lists E15 as final guidance dated April 2008 under docket FDA-2008-D-0199, issued by CDER and CBER. FDA states that consistent terminology supports integration of pharmacogenomics and pharmacogenetics into global drug development and approval. Preserve this guardrail: Scientific knowledge and regional privacy law have continued to develop, so current requirements also matter.
Try this
Draw the coding architecture for the april 2008 fda guidance is final. Record guidance status, date, docket, issuing centres and the present regional governance context. Show keys, access and participant actions. Respect this boundary: Scientific knowledge and regional privacy law have continued to develop, so current requirements also matter.
A strong science note separates observation from interpretation. Record guidance status, date, docket, issuing centres and the present regional governance context. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Record guidance status, date, docket, issuing centres and the present regional governance context. Keep this boundary visible: Scientific knowledge and regional privacy law have continued to develop, so current requirements also matter.
CHAPTER 3 OF 17 · Define the science
3. A genomic biomarker is a measurable DNA or RNA characteristic
E15 defines a genomic biomarker as a measurable DNA and/or RNA characteristic indicating normal biological processes, pathogenic processes or response to therapeutic or other interventions. It may comprise one or many characteristics.
A responsible explanation of a genomic biomarker is a measurable dna or rna characteristic connects laboratory science with traceability and consent. E15 defines a genomic biomarker as a measurable DNA and/or RNA characteristic indicating normal biological processes, pathogenic processes or response to therapeutic or other interventions. It may comprise one or many characteristics. Remember: The definition does not by itself show that a biomarker is analytically valid or fit for a proposed use.
Try this
Audit the consent-to-system match for a genomic biomarker is a measurable dna or rna characteristic. Write the characteristic, measurement method, biological interpretation and intended context separately. Find any promised action the coding design cannot deliver. Preserve this caution: The definition does not by itself show that a biomarker is analytically valid or fit for a proposed use.
A strong science note separates observation from interpretation. Write the characteristic, measurement method, biological interpretation and intended context separately. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Write the characteristic, measurement method, biological interpretation and intended context separately. Keep this boundary visible: The definition does not by itself show that a biomarker is analytically valid or fit for a proposed use.
Examples include single-nucleotide polymorphisms, repeat variation, haplotypes, methylation, insertions, deletions, copy-number changes and cytogenetic rearrangements. The category is broader than a single inherited sequence variant.
The clarity of dna characteristics take many forms begins with naming the biological measurement and the data relationship precisely. Examples include single-nucleotide polymorphisms, repeat variation, haplotypes, methylation, insertions, deletions, copy-number changes and cytogenetic rearrangements. The category is broader than a single inherited sequence variant. Keep this boundary visible: Different assays and tissues can measure different biological phenomena under the same broad label.
Try this
Classify a biomarker study for dna characteristics take many forms. Specify variant type, specimen, assay, quality controls and reference genome. Separate analyte, response and intended use. Keep this limit: Different assays and tissues can measure different biological phenomena under the same broad label.
A strong science note separates observation from interpretation. Specify variant type, specimen, assay, quality controls and reference genome. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Specify variant type, specimen, assay, quality controls and reference genome. Keep this boundary visible: Different assays and tissues can measure different biological phenomena under the same broad label.
CHAPTER 5 OF 17 · Connect genes and response
5. RNA characteristics add dynamic information
RNA sequences, expression levels, splicing, editing and microRNA levels can all serve as genomic characteristics under E15. These measures can change with tissue, disease, treatment and sample handling.
For rna characteristics add dynamic information, added privacy can remove future scientific or participant options. RNA sequences, expression levels, splicing, editing and microRNA levels can all serve as genomic characteristics under E15. These measures can change with tissue, disease, treatment and sample handling. Preserve this guardrail: An expression result is not automatically a stable inherited trait.
Try this
Draw the coding architecture for rna characteristics add dynamic information. Record tissue, collection time, preservation, assay and relevant biological context. Show keys, access and participant actions. Respect this boundary: An expression result is not automatically a stable inherited trait.
A strong science note separates observation from interpretation. Record tissue, collection time, preservation, assay and relevant biological context. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Record tissue, collection time, preservation, assay and relevant biological context. Keep this boundary visible: An expression result is not automatically a stable inherited trait.
CHAPTER 6 OF 17 · Connect genes and response
6. Genomic biomarkers are not restricted to humans
E15’s definition can include viral, infectious-agent and animal samples, supporting nonclinical and toxicological applications as well as human research. The source organism and intended inference should be explicit.
A responsible explanation of genomic biomarkers are not restricted to humans connects laboratory science with traceability and consent. E15’s definition can include viral, infectious-agent and animal samples, supporting nonclinical and toxicological applications as well as human research. The source organism and intended inference should be explicit. Remember: A microbial or animal marker should not be described as a human genomic characteristic without clarity.
Try this
Audit the consent-to-system match for genomic biomarkers are not restricted to humans. Label organism, specimen and translational purpose for every biomarker. Find any promised action the coding design cannot deliver. Preserve this caution: A microbial or animal marker should not be described as a human genomic characteristic without clarity.
A strong science note separates observation from interpretation. Label organism, specimen and translational purpose for every biomarker. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Label organism, specimen and translational purpose for every biomarker. Keep this boundary visible: A microbial or animal marker should not be described as a human genomic characteristic without clarity.
CHAPTER 7 OF 17 · Connect genes and response
7. Proteins and metabolites fall outside the E15 genomic definition
The E15 genomic-biomarker definition focuses on DNA and RNA and excludes measurement and characterisation of proteins or low-molecular-weight metabolites. Related omics fields can still be important but need their own terminology and evidence.
The clarity of proteins and metabolites fall outside the e15 genomic definition begins with naming the biological measurement and the data relationship precisely. The E15 genomic-biomarker definition focuses on DNA and RNA and excludes measurement and characterisation of proteins or low-molecular-weight metabolites. Related omics fields can still be important but need their own terminology and evidence. Keep this boundary visible: Calling every molecular measurement genomic blurs distinct methods and claims.
Try this
Classify a biomarker study for proteins and metabolites fall outside the e15 genomic definition. Classify each marker by analyte before discussing its role. Separate analyte, response and intended use. Keep this limit: Calling every molecular measurement genomic blurs distinct methods and claims.
A strong science note separates observation from interpretation. Classify each marker by analyte before discussing its role. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Classify each marker by analyte before discussing its role. Keep this boundary visible: Calling every molecular measurement genomic blurs distinct methods and claims.
CHAPTER 8 OF 17 · Connect genes and response
8. Pharmacogenomics connects DNA and RNA variation with drug response
E15 defines pharmacogenomics as the study of variations of DNA and RNA characteristics as related to drug response. Drug response includes pharmacokinetics, pharmacodynamics, efficacy and adverse effects.
For pharmacogenomics connects dna and rna variation with drug response, added privacy can remove future scientific or participant options. E15 defines pharmacogenomics as the study of variations of DNA and RNA characteristics as related to drug response. Drug response includes pharmacokinetics, pharmacodynamics, efficacy and adverse effects. Preserve this guardrail: Association does not automatically imply clinical utility or a treatment recommendation.
Try this
Draw the coding architecture for pharmacogenomics connects dna and rna variation with drug response. State the response phenotype, population, design, effect estimate and intended decision. Show keys, access and participant actions. Respect this boundary: Association does not automatically imply clinical utility or a treatment recommendation.
A strong science note separates observation from interpretation. State the response phenotype, population, design, effect estimate and intended decision. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
State the response phenotype, population, design, effect estimate and intended decision. Keep this boundary visible: Association does not automatically imply clinical utility or a treatment recommendation.
CHAPTER 9 OF 17 · Choose the coding
9. Pharmacogenetics is a subset of pharmacogenomics
Pharmacogenetics focuses on variation in DNA sequence as related to drug response. The subset relation helps explain why an RNA-expression study may be pharmacogenomic without being pharmacogenetic.
A responsible explanation of pharmacogenetics is a subset of pharmacogenomics connects laboratory science with traceability and consent. Pharmacogenetics focuses on variation in DNA sequence as related to drug response. The subset relation helps explain why an RNA-expression study may be pharmacogenomic without being pharmacogenetic. Remember: The terms should not be used interchangeably when the biological distinction matters.
Try this
Audit the consent-to-system match for pharmacogenetics is a subset of pharmacogenomics. Classify example studies and justify PGx, PGt or another field. Find any promised action the coding design cannot deliver. Preserve this caution: The terms should not be used interchangeably when the biological distinction matters.
A strong science note separates observation from interpretation. Classify example studies and justify PGx, PGt or another field. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Classify example studies and justify PGx, PGt or another field. Keep this boundary visible: The terms should not be used interchangeably when the biological distinction matters.
CHAPTER 10 OF 17 · Choose the coding
10. The medicine concept includes biological products and vaccines
E15 uses drug broadly to cover investigational and marketed medicinal or pharmaceutical products, including vaccines and other biological products. The definitions therefore travel across many development settings.
The clarity of the medicine concept includes biological products and vaccines begins with naming the biological measurement and the data relationship precisely. E15 uses drug broadly to cover investigational and marketed medicinal or pharmaceutical products, including vaccines and other biological products. The definitions therefore travel across many development settings. Keep this boundary visible: Product-specific biology and regulatory context still affect evidence requirements.
Try this
Classify a biomarker study for the medicine concept includes biological products and vaccines. Record product modality and why the biomarker question is relevant. Separate analyte, response and intended use. Keep this limit: Product-specific biology and regulatory context still affect evidence requirements.
A strong science note separates observation from interpretation. Record product modality and why the biomarker question is relevant. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Record product modality and why the biomarker question is relevant. Keep this boundary visible: Product-specific biology and regulatory context still affect evidence requirements.
Identified data and samples carry personal identifiers and can support monitoring, follow-up, adding new data, withdrawal or return of results. E15 says they are generally not considered appropriate for drug-development clinical trials.
For identified samples are directly traceable, added privacy can remove future scientific or participant options. Identified data and samples carry personal identifiers and can support monitoring, follow-up, adding new data, withdrawal or return of results. E15 says they are generally not considered appropriate for drug-development clinical trials. Preserve this guardrail: Direct traceability increases privacy risk and requires strong justification and controls.
Try this
Draw the coding architecture for identified samples are directly traceable. Map identifiers, access, purpose and safer coding alternatives. Show keys, access and participant actions. Respect this boundary: Direct traceability increases privacy risk and requires strong justification and controls.
A strong science note separates observation from interpretation. Map identifiers, access, purpose and safer coding alternatives. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Map identifiers, access, purpose and safer coding alternatives. Keep this boundary visible: Direct traceability increases privacy risk and requires strong justification and controls.
Single-coded materials carry a code rather than personal identifiers, with one key allowing indirect traceability. The model supports follow-up and participant-related actions while adding protection compared with directly identified data.
A responsible explanation of single-coded samples use one key connects laboratory science with traceability and consent. Single-coded materials carry a code rather than personal identifiers, with one key allowing indirect traceability. The model supports follow-up and participant-related actions while adding protection compared with directly identified data. Remember: Anyone with code and key access may reconnect the sample to the participant.
Try this
Audit the consent-to-system match for single-coded samples use one key. Separate sample handling from key custody and audit every access. Find any promised action the coding design cannot deliver. Preserve this caution: Anyone with code and key access may reconnect the sample to the participant.
A strong science note separates observation from interpretation. Separate sample handling from key custody and audit every access. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Separate sample handling from key custody and audit every access. Keep this boundary visible: Anyone with code and key access may reconnect the sample to the participant.
CHAPTER 13 OF 17 · Protect future meaning
13. Double-coded samples add another barrier
Double coding relabels the material with a second code linked through a second key. Access to both keys is needed to reconnect the sample, while follow-up or new data may still be possible under controlled arrangements.
The clarity of double-coded samples add another barrier begins with naming the biological measurement and the data relationship precisely. Double coding relabels the material with a second code linked through a second key. Access to both keys is needed to reconnect the sample, while follow-up or new data may still be possible under controlled arrangements. Keep this boundary visible: Two codes provide additional protection only if key roles and systems are genuinely separated.
Try this
Classify a biomarker study for double-coded samples add another barrier. Draw the two-key flow and test who can re-identify under normal and emergency conditions. Separate analyte, response and intended use. Keep this limit: Two codes provide additional protection only if key roles and systems are genuinely separated.
A strong science note separates observation from interpretation. Draw the two-key flow and test who can re-identify under normal and emergency conditions. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Draw the two-key flow and test who can re-identify under normal and emergency conditions. Keep this boundary visible: Two codes provide additional protection only if key roles and systems are genuinely separated.
Anonymised materials were initially coded, but the link to personal identifiers is deleted. E15 explains that withdrawal, return of individual results, clinical monitoring, follow-up and adding new participant data are then no longer possible through those keys.
For anonymised samples lose the code link, added privacy can remove future scientific or participant options. Anonymised materials were initially coded, but the link to personal identifiers is deleted. E15 explains that withdrawal, return of individual results, clinical monitoring, follow-up and adding new participant data are then no longer possible through those keys. Preserve this guardrail: Deleting the key trades future traceability and participant choice for additional protection.
Try this
Draw the coding architecture for anonymised samples lose the code link. Document the deletion authority, timing, verification and scientific consequences. Show keys, access and participant actions. Respect this boundary: Deleting the key trades future traceability and participant choice for additional protection.
A strong science note separates observation from interpretation. Document the deletion authority, timing, verification and scientific consequences. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Document the deletion authority, timing, verification and scientific consequences. Keep this boundary visible: Deleting the key trades future traceability and participant choice for additional protection.
CHAPTER 15 OF 17 · Protect future meaning
15. Anonymous samples never had a coding key
Anonymous samples are collected without personal identifiers and without creation of a coding key. They cannot be traced to an individual, and only limited clinical information may accompany them.
A responsible explanation of anonymous samples never had a coding key connects laboratory science with traceability and consent. Anonymous samples are collected without personal identifiers and without creation of a coding key. They cannot be traced to an individual, and only limited clinical information may accompany them. Remember: Anonymous and anonymised describe different histories and should not be conflated.
Try this
Audit the consent-to-system match for anonymous samples never had a coding key. Ask whether identifiers or a key ever existed before assigning the category. Find any promised action the coding design cannot deliver. Preserve this caution: Anonymous and anonymised describe different histories and should not be conflated.
A strong science note separates observation from interpretation. Ask whether identifiers or a key ever existed before assigning the category. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Ask whether identifiers or a key ever existed before assigning the category. Keep this boundary visible: Anonymous and anonymised describe different histories and should not be conflated.
CHAPTER 16 OF 17 · Protect future meaning
16. Coding choice shapes consent and scientific options
Traceability affects withdrawal, return of results, clinical follow-up and addition of new data. E15 says the conditions for linking back, including return of genomic data, should be described in research documents such as informed consent.
The clarity of coding choice shapes consent and scientific options begins with naming the biological measurement and the data relationship precisely. Traceability affects withdrawal, return of results, clinical follow-up and addition of new data. E15 says the conditions for linking back, including return of genomic data, should be described in research documents such as informed consent. Keep this boundary visible: E15 does not itself resolve all consent, privacy or return-of-results requirements.
Try this
Classify a biomarker study for coding choice shapes consent and scientific options. Align consent language, coding architecture and operational capability before collection. Separate analyte, response and intended use. Keep this limit: E15 does not itself resolve all consent, privacy or return-of-results requirements.
A strong science note separates observation from interpretation. Align consent language, coding architecture and operational capability before collection. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Align consent language, coding architecture and operational capability before collection. Keep this boundary visible: E15 does not itself resolve all consent, privacy or return-of-results requirements.
CHAPTER 17 OF 17 · Protect future meaning
17. A complete E15 receipt connects terminology to governance
The receipt records analyte, biomarker definition, PGx or PGt classification, product response, specimen source, assay, coding category, key custody, traceability, withdrawal, return of results, consent, privacy, retention and future-use limits. It makes the design intelligible.
For a complete e15 receipt connects terminology to governance, added privacy can remove future scientific or participant options. The receipt records analyte, biomarker definition, PGx or PGt classification, product response, specimen source, assay, coding category, key custody, traceability, withdrawal, return of results, consent, privacy, retention and future-use limits. It makes the design intelligible. Preserve this guardrail: A precise label cannot compensate for weak consent, insecure systems or an unvalidated assay.
Try this
Draw the coding architecture for a complete e15 receipt connects terminology to governance. Give the receipt to biology, data-protection and participant-facing reviewers and reconcile their assumptions. Show keys, access and participant actions. Respect this boundary: A precise label cannot compensate for weak consent, insecure systems or an unvalidated assay.
A strong science note separates observation from interpretation. Give the receipt to biology, data-protection and participant-facing reviewers and reconcile their assumptions. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.
Evidence checkpoint
Give the receipt to biology, data-protection and participant-facing reviewers and reconcile their assumptions. Keep this boundary visible: A precise label cannot compensate for weak consent, insecure systems or an unvalidated assay.
E15 genomic terminology and coding map
| Question | Evidence to capture | Boundary to retain |
|---|---|---|
| What is measured? | DNA or RNA characteristic and assay | Definition is not validation |
| What is studied? | Variation related to drug response | Association is not clinical utility |
| How is it linked? | Identified, single- or double-coded | Key control determines traceability |
| Can it reconnect? | Anonymised or anonymous history | Privacy changes future options |
Science learning becomes useful when a student can show where a claim came from, why it applies and where it stops. Continue with the eduKate Science Learning Hub, compare choices in the Education Hub, and connect these habits to career pathways built from subject capabilities.
E15 shows that precise words are practical tools. When scientists distinguish the biomarker, response and coding history clearly, they can design better studies, explain choices honestly and protect both participants and the future value of genomic research.
Primary sources and verification date
Source status checked 26 October 2026. FDA guidance describes the agency’s current thinking and does not replace applicable statutes, regulations or product-specific communication with FDA.
- FDA: ICH E15 Pharmacogenomics Definitions and Sample Coding — Final guidance, April 2008; docket FDA-2008-D-0199.
- FDA: Full ICH E15 guidance PDF — Official genomic definitions and four sample-coding categories.
- FDA: Clinical Pharmacogenomics in Early-Phase Studies and Labeling — FDA recommendations for considering genomic variation in drug development.