The study introduces a double-indexing method to improve the accuracy of multiplex sequencing on the Illumina platform. Traditional single-indexing strategies, which embed indexes in one adapter, can lead to sample misidentification due to mixed clusters, cross-contamination, and PCR jumping. The double-indexing method places indexes in both adapters, allowing for independent verification of sample identity. The researchers found that the false-assignment rate of samples was significantly higher than expected, reaching up to 0.3%, primarily due to mixed clusters. They identified two additional sources of error: sporadic cross-contamination of oligonucleotides and PCR jumping during bulk amplification. The study suggests that double-indexing can reduce false-assignment rates by an order of magnitude, making it suitable for applications requiring high accuracy, such as cancer genomics and ancient DNA research. The method also reduces the cost of highly multiplexed sequencing by allowing for a larger number of index combinations.The study introduces a double-indexing method to improve the accuracy of multiplex sequencing on the Illumina platform. Traditional single-indexing strategies, which embed indexes in one adapter, can lead to sample misidentification due to mixed clusters, cross-contamination, and PCR jumping. The double-indexing method places indexes in both adapters, allowing for independent verification of sample identity. The researchers found that the false-assignment rate of samples was significantly higher than expected, reaching up to 0.3%, primarily due to mixed clusters. They identified two additional sources of error: sporadic cross-contamination of oligonucleotides and PCR jumping during bulk amplification. The study suggests that double-indexing can reduce false-assignment rates by an order of magnitude, making it suitable for applications requiring high accuracy, such as cancer genomics and ancient DNA research. The method also reduces the cost of highly multiplexed sequencing by allowing for a larger number of index combinations.