Transform Data Table - Extract Data from Existing Column and Return New Column

you can use Merge for that, merging with current table, with following joins (you add second and more joins by clicking on a column with Ctrl pressed):

you can then expand the relevant column and rename it properly

you can also see how it works with this M query

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("Pc3LDcAwCAPQXThzKKH5zYKy/xoB6nJwFD1Z2IyEuLI8jQ6bv5wJHaXqP4XO0heS/ScrwR2ULMUDN4J36cQNxZ6kLvQUe59uiP57zucC", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type text) meta [Serialized.Text = true]) in type table [Index = _t, ATD = _t, #"run id" = _t, #"data 1" = _t, #"Ref id" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Index", Int64.Type}, {"ATD", Int64.Type}, {"run id", Int64.Type}, {"data 1", Int64.Type}, {"Ref id", Int64.Type}}),
    #"Merged Queries" = Table.NestedJoin(#"Changed Type", {"Ref id", "ATD"}, #"Changed Type", {"run id", "ATD"}, "Changed Type", JoinKind.LeftOuter),
    #"Expanded Changed Type" = Table.ExpandTableColumn(#"Merged Queries", "Changed Type", {"data 1"}, {"Changed Type.data 1"}),
    #"Renamed Columns" = Table.RenameColumns(#"Expanded Changed Type",{{"Changed Type.data 1", "Ref data"}})
in
    #"Renamed Columns"