Untangle improperly sorted columns

untangle.py

lines = """\
ABCDE12345_001,FGHIJ6789_002,ABCDE12345_001
KLMNO5432_003,KLMNO5432_003,FGHIJ6789_002
PQRST24680_123,UVWXY13579_555,UVWXY13579_555
ZABCD876530_009,ZABCD876530_009,AABBCCDDEE_987
AABBCCDDEE_987,AABBCCDDEE_987,LMNOP98765_999
LMNOP98765_999,ZYXWV54321_777,ZYXWV54321_777""".splitlines()

record = []
for row in lines:
    record += [(word,ordinal) for ordinal,word in enumerate(row.split(','))]

record.sort()

last = None
build = None
for word,posn in record:
    if word != last:
        if build:
            print(','.join(build))
        build = ['','','']
        last = word
    build[posn] = word

print(','.join(build))

Note that I fixed the lone "7" in the 4th data line, and removed an extra "," in the 3rd data line.