Any plans to implement anchor binding position in pVACseq

[immuneAI]

Hello,

Is there any plans to implement this workflow https://github.com/griffithlab/anchor_huiming_etal_2023 in pvacseq, described in the https://www.science.org/doi/full/10.1126/sciimmunol.abg2200 manuscript. This would be a great help and solve key issues in neoantigens binding.

Much appreciated your help in keeping the tool up to date and support!!

Best, Venkat

[susannasiebert]

Hi Venkat,

We currently use the anchor criteria in a couple of ways in pVACtools.

• In the aggregate report we use it to deprioritize neoepitope candidates that fail this criteria (see https://pvactools.readthedocs.io/en/latest/pvacseq/output_files.html#best-peptide-criteria)
• If none of the candidates pass the anchor criteria, the variant gets tiered into the "Anchor" tier (see https://pvactools.readthedocs.io/en/latest/pvacseq/output_files.html#the-pvacseq-aggregate-report-tiers)
• In pVACview we display heatmaps of the anchor probabilities for each amino acid of any candidate passing the anchor-constribution-threshold (see example at the bottom of the page here: https://pvactools.readthedocs.io/en/latest/pvacview/features.html#additional-peptide-information-additional-information-regarding-individual-algorithm-binding-and-anchor-scores) and show whether each candidate passed or failed the anchor criteria.

Are there any additional ways you would like us to incorporate this information?

[susannasiebert]

Closing this issue due to inactivity.