Recursive generation of partitionable graphs

Results of Lovász (1972) and Padberg (1974) imply that partitionable graphs contain all the potential counterexamples to Berge’s famous Strong Perfect Graph Conjecture. A recursive method of generating partitionable graphs was suggested by Chvátal, Graham, Perold and Whitesides (1979). Results of Sebő (1996) entail that Berge’s conjecture holds for all the partitionable graphs obtained by this method. Here we suggest a more general recursion. Computer experiments show that it generates all the partitionable graphs with ω = 3, α ≤ 9 (and we conjecture that the same will hold for bigger α, too) and many but not all for (ω, α) = (4, 4) and (4, 5). Here α and ω are respectively the clique and stability numbers of a partitionable graph, that is the numbers of vertices in its maximum cliques and stable sets. All the partitionable graphs generated by our method contain a critical ω-clique, that is an ω-clique which intersects only 2ω − 2 other ω-cliques. This property might imply that in our class there are no counterexamples to Berge’s conjecture (cf. Sebő (1996)), however this question is still open.