# Sequencing reads

Source: https://onco.cc/terms/sequencing-reads/  
OnCo record `sequencing-reads` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

A sequencing read records base letters from prepared material, with estimates of their quality.

## Summary

A read is an instrument-derived base sequence from a prepared library molecule with per-base quality; library preparation can copy, shorten or select molecules before sequencing.

## Fields

- Kind: Term
- Last checked: 2026-11-15

## Notes

- A read is an instrument-derived base sequence from a prepared library molecule with per-base quality; library preparation can copy, shorten or select molecules before sequencing.
- Phred quality: Q20 is an estimated 1% per-base error probability and Q30 0.1%; these are model-based base-call estimates, not the probability that a reported mutation is real; Q30 is a scale value, not a universal pass rule.
- Base quality (a letter) and mapping confidence (a location) are separate; repeats can make placement ambiguous; one read can have more than one proposed alignment record.
- Paired-end reads come from opposite ends of one fragment, so they are complementary placement evidence, not two independent molecules.
- Amplification copies one fragment many times; duplicate recognition or validated molecular tags assess independent support but do not repair errors introduced before tagging; in amplicon assays identical start and end coordinates cannot separate copies from independent molecules.
- Raw versus usable reads: filtering changes the denominator for downstream support; a read does not identify a cell, establish function or alone prove a somatic mutation.
- Record extraction and library preparation, specimen compatibility, read lengths, base qualities, pairs, alignment records and molecule-support rules. Preparation consumes material; reanalysis does not. Contamination, damaged molecules, adapters, amplification and mapping ambiguity affect support. Filter settings and Q30 are not universal clinical acceptance criteria; validate the complete specimen-to-report workflow.
- In one fictional example, ten paired-end reads form five pairs, representing at most five library fragments; amplification copies can reduce independent source support further. In a separate example, one source fragment has three sequenced copies, each yielding two reads, for six reads from that one ancestor. Do not add the counts from these two examples. Ask how the preparation and analysis established molecular independence.

## Sources

- Guidelines for Validation of Next-Generation Sequencing-Based Oncology Panels: A Joint Consensus Recommendation of the Association for Molecular Pathology and College of American Pathologists.: https://pubmed.ncbi.nlm.nih.gov/28341590/
- Base-calling of automated sequencer traces using phred. II. Error probabilities.: https://pubmed.ncbi.nlm.nih.gov/9521922/
- SAM specification: https://samtools.github.io/hts-specs/SAMv1.pdf
- NCBI SRA: FASTQ records and paired-end submission formats: https://www.ncbi.nlm.nih.gov/sra/docs/submitformats/

## Connected records

- terms: [Next-generation sequencing (NGS)](https://onco.cc/terms/ngs/)

---
JSON: https://onco.cc/api/v1/entities/sequencing-reads.json