Vaccine combinations

test.scala

val df = spark.read.option("header", true)
  .csv("/Users/elena/Downloads/vaccine_combinations.csv")

df.createTempView("data")

val diseases = df.columns.filter(_ != "ID")

diseases.map(d => df.where(col(d) === lit(1)).select(col("ID"), lit(d).as("disease")))
  .reduce(_ union _)
  .createTempView("vac2dis")

spark.sql(
  """select count(*)
    |from data as v1
    |join data as v2 on v1.ID < v2.ID
    |where not exists (
    |  select 1
    |  from vac2dis d1
    |  join vac2dis d2
    |    on d1.disease = d2.disease
    |  where d1.ID = v1.ID and d2.ID = v2.ID
    |)
    |""".stripMargin
).show()

The dataset can be found here.