Illumina high-throughput sequencing has transformed the world of genomics by allowing for simultaneous analysis of many samples in a single run. This relies on sample indexes, which are short DNA sequences added during library preparation allowing the reads to be assigned back to the original sample. While this process is well-established, it has led to the phenomenon known as index hopping. Knowing what index hopping is, and how to mitigate it, is key for generating reliable sequencing data.
Index hopping, sometimes called index switching, occurs when a sequencing read is assigned to the wrong sample index. As a result, reads originating from one library are incorrectly attributed to another library that was pooled and sequenced in the same run. This creates low level cross sample contamination that can be difficult to detect without appropriate controls.
The impact of index hopping depends on the application. In experiments with large differences in library complexity or abundance, even a small fraction of misassigned reads can affect results. This is particularly relevant for applications such as low frequency variant detection, single cell sequencing, and metagenomics, where distinguishing true signal from background noise is critical.
Typically, index hopping is due to the presence of unpurified free adapters or index primers in a pooled library. The free adapters or primers can anneal to DNA fragments from other samples. During amplification or cluster generation, they can then be incorporated into a fragment, effectively replacing the original index.
It is important to note that index hopping is distinct from barcode cross talk caused by sequencing errors. In index hopping, the index sequence itself is physically swapped, meaning the index is read accurately but does not match the original sample.
Not all sequencing runs are equally affected by index hopping. Libraries with very high read counts can contribute more hopped reads, simply due to their abundance. When such libraries are pooled with low input or low complexity samples, the relative impact becomes more pronounced.
Another contributing factor is the use of single-indexed libraries. When there is only one index present, no secondary identifier exists to confirm the sample identification. If that single index is replaced or misread, the read will be confidently, but incorrectly, assigned to the wrong sample.
Additionally, applications that rely on detecting rare events are especially sensitive. For example, in tumor sequencing, misassigned reads from a high-coverage normal sample could obscure or mimic low frequency variants in a tumor sample. In single cell experiments, even a small number of contaminating reads can affect cell classification or gene expression estimates.
One of the most effective strategies to mitigate index hopping is the use of unique dual indexes. In this design, each library carries two independent index sequences. If index hopping occurs, it is highly unlikely that both indexes will be replaced in a coordinated way that matches another valid index pair. As a result, hopped reads are easily identified and discarded during demultiplexing. Unique dual indexing has become a standard best practice for applications that demand high accuracy.
Improvements in library cleanup also play a role: more efficient removal of free adapters and primers reduces the pool of index sequences to potentially hop. Bead based size selection and optimized cleanup protocols help ensure that only fully constructed libraries are carried forward to sequencing. Assessing adapter and primer presence using a Bioanalyzer or TapeStation can help determine whether additional clean‑up steps are needed.
Sequencing platform chemistry and workflow improvements have reduced the impact of index hopping. Modern exclusion amplification chemistries limit the opportunity for free adapters to interact with library fragments during cluster generation, while updated flow cell designs and reagents are optimized to minimize index swapping events.
Additionally, improved demultiplexing algorithms allow for stricter index matching and better filtering of unexpected index combinations. When combined with dual indexing, these computational safeguards provide another layer of protection against misassignment.
Index hopping is a well characterized artifact of multiplexed sequencing that results from the unintended reassignment of sample indexes. Although it can introduce cross sample contamination, advances in library design and sequencing platform chemistry have significantly reduced its impact. By understanding how index hopping occurs and by adopting best practices such as unique dual indexing and adapter/primer cleanup, researchers can generate high quality sequencing data with confidence.