Molecular Barcode

A molecular barcode is a random or semi-random synthetic nucleotide sequence (typically 8–16 base pairs) ligated to individual DNA fragments during library preparation. The key design principles include:

• Diversity: A barcode length of N bases yields 4^N possible unique sequences, ensuring each original molecule in a sample receives a distinct tag.
• Hamming Distance: Barcodes are designed with sufficient edit distance to prevent misassignment due to sequencing errors within the barcode itself.
• Duplex Tagging: In advanced protocols, complementary barcodes are attached to both strands of a DNA duplex, enabling true double-strand consensus and distinguishing real mutations from single-strand damage.

After sequencing, reads sharing the same molecular barcode are grouped into read families. A consensus sequence is computationally derived by comparing all reads within a family:

• Single-Strand Consensus (SSCS): Redundant reads from one strand are collapsed to eliminate random polymerase errors.
• Duplex Consensus (DCS): SSCS reads from both the forward and reverse strands are compared. A true mutation must be present on both strands, effectively eliminating false positives from oxidative damage or cytosine deamination.
• Error Rate Suppression: This process reduces the effective error rate from ~1% (raw sequencing) to as low as 10^-7 to 10^-8.

Unlike standard sequencing, which measures relative abundance, molecular barcodes enable digital counting of input molecules. By counting the number of unique barcode families rather than total sequencing reads, the system corrects for PCR amplification bias:

• Input Molecule Count: The number of distinct barcodes recovered is directly proportional to the number of original DNA fragments.
• Variant Allele Frequency (VAF): True VAF is calculated as (mutant barcode families) / (total barcode families), not by read depth.
• Library Complexity: The ratio of unique barcodes to total reads serves as a quality control metric for library diversity and over-amplification.

Molecular barcodes computationally separate biological signal from three distinct error sources:

• Polymerase Errors: Random misincorporations during PCR amplification are diluted out by requiring multiple identical copies within a read family.
• Sequencer Base-Calling Errors: Stochastic errors on the flow cell are suppressed by consensus, as the probability of the same random error occurring in multiple reads of the same family is negligible.
• Base Damage Artifacts: Cytosine deamination (C>T) and guanine oxidation (G>T) are strand-specific events. Duplex consensus requires the lesion to be present on both strands, effectively filtering these artifacts.

Molecular barcodes provide a secondary line of defense against sample cross-contamination caused by index hopping. When free-floating index primers re-associate with the wrong template during cluster amplification, the molecular barcode remains physically ligated to the original fragment:

• Contamination Detection: A read with a sample index for Patient A but a molecular barcode from Patient B's library can be flagged and removed.
• In-line Barcoding: Some protocols embed the sample identifier within the molecular barcode sequence itself, creating a unified tag that cannot be reassigned.

Duplex sequencing represents the gold standard of molecular barcoding, where both strands of the original DNA duplex are independently tagged, amplified, and sequenced. The process:

• Strand Pairing: After sequencing, the complementary barcodes on the Watson and Crick strands are computationally reunited.
• True Mutation Confirmation: A variant is only called if it appears at the same position in both independent strand consensus sequences.
• Limit of Detection (LoD): This method achieves an LoD below 1 in 10,000 molecules, making it suitable for detecting minimal residual disease where tumor fraction may be vanishingly small.

A random nucleotide barcode ligated to individual DNA molecules before amplification, enabling computational deduplication and absolute quantification of original template molecules. UMIs are the direct implementation of molecular barcoding, typically consisting of 8-12 degenerate bases that create a unique tag for each starting fragment. During analysis, reads sharing the same UMI are collapsed into a consensus sequence, eliminating PCR duplicates and suppressing polymerase errors. This allows precise counting of input molecules rather than amplified copies, critical for detecting variants at allele frequencies below 1%.

An error-correction method that independently sequences both strands of a DNA duplex using complementary UMIs to distinguish true mutations from polymerase errors and base damage. Each strand receives a distinct UMI, and true variants must be present on both strands at the same position. This dual-strand consensus achieves error rates as low as 1 in 10^7 bases, compared to ~1 in 1,000 for standard sequencing. Duplex sequencing is the gold standard for detecting ultra-rare variants in liquid biopsy applications where false positives from oxidative damage or deamination must be eliminated.

A bioinformatic strategy that leverages molecular barcodes and redundant sequencing to build consensus sequences, suppressing random polymerase and sequencer errors below the variant detection threshold. The process involves three steps:

• Grouping: Reads are clustered by shared UMI sequences
• Alignment: Reads within each UMI family are aligned to identify discordant bases
• Consensus calling: A majority-vote or probabilistic model determines the true base at each position This approach converts stochastic errors into systematic noise that can be computationally removed, enabling reliable detection of variants present in as few as 0.01% of input molecules.

The number of unique, non-duplicate DNA molecules in a sequencing library, reflecting the diversity of the original input material and the quantitative power of the assay. Molecular barcodes enable direct measurement of library complexity by counting distinct UMI sequences after deduplication. High complexity indicates sufficient input DNA and minimal amplification bias, while low complexity suggests bottlenecking during library preparation. In liquid biopsy, library complexity directly determines the limit of detection—insufficient unique molecules mean rare tumor-derived fragments may be absent from the library entirely, producing false negatives regardless of sequencing depth.

A sequencing artifact where sample barcodes are misassigned due to index-switching during cluster amplification, causing sample cross-contamination that must be computationally mitigated. Index hopping occurs when free-floating index primers in the flow cell anneal to the wrong template during bridge amplification. Molecular barcodes provide a secondary layer of sample identification—even if index hopping assigns a read to the wrong sample, the UMI sequence can flag the mismatch. Dual-indexing combined with UMIs creates a robust combinatorial identification system that reduces sample misassignment rates to negligible levels.

The percentage of sequencing reads at a specific genomic locus that contain a variant allele, used to estimate the proportion of mutated DNA molecules in a heterogeneous sample. Molecular barcodes transform VAF from a relative abundance estimate distorted by PCR amplification bias into an absolute molecular count. Without UMIs, over-amplified fragments inflate apparent variant frequency; with UMIs, each unique molecule contributes exactly one count to the VAF calculation. This molecular VAF enables accurate monitoring of tumor burden changes over time, with changes as small as 0.1% being clinically meaningful for minimal residual disease tracking.