This document outlines the steps required to design a targeted single-cell CpG methylation analysis panel based on the original scTAM-seq assay (Ref 1). In principle, methylation panel design can be broken down into these steps.

1. Selection of the restriction enzyme. The choice of the restriction enzyme dictates the number and type of methylation sites that can be analyzed. We recommend HhaI; it was used in both scTAM-seq and EPI-Clone publications (Ref 1, 2), and is fully compatible with Tapestri chemistry.
2. Selection of the target regions. This step aims to generate the list of genomic regions that will serve as targets for panel design. The targets are biologically informative CpG sites that satisfy the requirements of the chosen restriction enzyme.
3. Design of a targeted panel The targets are submitted to Tapestri Designer, which aims to create amplicons to cover those targets. 

Selection of Target Regions

Once the decision on the enzyme has been made, the next step is to define genomic regions that will be targeted by the panel. Two sets of criteria are used in this process - genomic context of a target region and biological usefulness of CpG sites.

Genomics context can be defined in the following way:

1. Only one enzyme recognition site in a 300 bp window (except for the control amplicons without the recognition site; see below)
2. GC content between 45% and 65% (note that this range is slightly different than standard recommendations for Tapestri Designer; see below). The upper end of the acceptable %GC range is especially important for promoter regions, where the cut off should be lower (~60%)

To identify biologically meaningful targets (the ability of a target to serve as a "methylation signature”, e.g. to differentiate between different cell states or populations), bulk methylation experiments should ideally be performed up front (e.g., methylation arrays or sequencing) to identify CpG sites of interest. If experimental data are not available, then a literature search or a database search should be conducted (suggested databases: IHEC, BLUEPRINT, or ENCODE).

The final list of targets should include:

1. Differentially methylated CpGs of interest with the enzyme recognition site.  The actual number depends on the experimental goal, but as a rule of thumb, we recommend 30 or more CpGs per cell state/cluster for projects aimed at distinguishing cell differentiation states (for example, to delineate six differentiation states, aim for 180 or more CpGs).
2. Control targets without the enzyme recognition site. These amplicons are used to distinguish between cells and non-cell barcodes. We recommend including 50 amplicons without the enzyme recognition site.  The list of 50 control amplicons from the DNA CpG Methylation Heme Panel (Lars Velten, CRG) is attached below. 
3. Control targets with a constitutively unmethylated enzyme recognition site (30-40 amplicons). The loss of signal from these amplicons is used to assess efficiency of digestion.

Designing a Targeted Panel

The list of targets created above is then used as input for Tapestri Designer (note that some targets may be refused by Tapestri Designer and that the criteria for methylation panels are more stringent than for the standard DNA panels). Once the design is complete, we strongly recommend the following:

1. Amplicons with GC content over 65% should be excluded (Tapestri Designer flags amplicons with GC content greater than 62%).
2. A final check that each amplicon sequence contains at most one cut site.

References

1. Bianchi, A., Scherer, M., Zaurin, R. et al. scTAM-seq enables targeted high-confidence analysis of DNA methylation in single cells. Genome Biol 23, 229 (2022). https://doi.org/10.1186/s13059-022-02796-7
2. Scherer, M., Singh, I., Braun, M.M. et al. Clonal tracing with somatic epimutations reveals dynamics of blood ageing. Nature 643, 478–487 (2025). https://doi.org/10.1038/s41586-025-09041-8 

• Control Ampliconsxlsx.xlsx10 KB