Ideally, you would troubleshoot low CUT&Tag yields, repeat the experiment, and then proceed to sequencing. This will result in higher quality sequencing data compared to using ultra-low CUT&Tag yields. However, in some cases, repeating the experiment isn’t possible. In fact, for some targets and cell types, low CUT&Tag yields are unavoidable.
If it is not possible to repeat the experiment, use a Speedvac to increase library concentration and add as much of the library as possible to the sequencing pool. Deeper sequencing (> 8 million total reads) is recommended to ensure sufficient read depth and fully capture library diversity.
Read duplicates may be increased but can be filtered out with Picard (broadinstitute.github.io/picard).