I used FreeBayes to call variants from BAM file and the reference genome is contigs of RNA-Seq. I have depth of coverage of each variation position. I want to use depth of k (3, 4, 5...) prior and posterior neighbor positions of each variant position to normalize depth of that variant position using this formula:

depth of variant position * (sequence mean depth/ neighbors mean depth)

How can I include depth of coverage of k prior and posterior neighbors of a position in VCF file?

Notes: 1- I can achieve depth of each position of each sequence using Samtools depth, but I think there must be a better way to directly have mean neighbor positions depths in VCF file. 2- I use galaxy but command line suggestions are also appreciated. Thank you.