Short-read sequencing libraries are usually not sequenced individually. Due to individual, uniquely used indices and respective pooling opportunities, several samples can be sequenced in parallel on a single flow cell. A well-chosen pooling strategy can save time and money and, therefore, optimize your project.
What does ‘pooling’ mean? Pooling stands for mixing individual samples. In this article, the pooling of sequencing libraries is explained. We focus on what is important when mixing the sequencing libraries.
Pooling can only be realized by having individually barcoded libraries, as each barcode can only be present once in the sequencing pool. Barcoding itself is part of the library generation process. If non-barcoded samples are pooled, assigning a sequencing read to its individual starting sample is impossible. Moreover, all samples in a sequencing pool are sequenced with the same read length. Thus, both factors — barcoding and sequencing mode — need to be considered thoroughly.
If all pooling requirements are met, two common strategies are available: equimolar pooling and non-equimolar pooling. The difference between both strategies is the ratio of the samples present in the pool. For equimolar pooling, all samples are present in the same ratio, whereas samples are mixed in different ratios when applying non-equimolar pooling. One main factor determining whether your samples need to be pooled equimolar or non-equimolar is the ratio of the samples’ required output.
Let’s make it more precise using an example. Thus, please have a look at figure 1. We assume having three samples, which shall be sequenced in one pool. After the sequencing library generation, all libraries are present in different concentrations. Before the pooling step, the libraries are normalized, resulting in the same concentration for all three sequencing libraries. This intermediate step is not mandatory but can simplify the pool creation in the next step. Further, figure 1 shows how two sequencing pools are generated: one equimolar and the other non-equimolar. The equimolar pool contains the same proportions of each library, whereas the libraries are mixed in a ratio of 3:2:1 for the non-equimolar pool. Finally, reads are generated during the sequencing process for each sequencing library. These sequencing reads can then be assigned to their starting sample. The read distribution of the three libraries is equal for the equimolar pool. The distribution of the reads for the non-equimolar pool corresponds to the ratio of the libraries in the pool.
Figure 1 | Pooling strategy workflow and corresponding sequencing results.
Consequently, pooling influences the outcome of each sequencing project. The right strategy ultimately depends on how many indices are available, which flow cell the pool is sequenced on, and how high the desired output per sample is. We are aware of that for all our products. Pooling is our daily business and is worth considering when starting a sequencing project.