What steps will reproduce the problem?

wipe () read ("rand0.fas") set (root: "Seq1") transform ((all, tcm:(1,1)), gap_opening:1) build (branch_and_bound, trees:100) rediagnose () select () calculate_support (bremer, build(100), swap(trees:100, all)) report ("rand0.dynamicbremer", supports) report (supports) exit ()

What is the expected output?

Bremer Supports: Support tree: (Seq1 (Seq2 (Seq3 Seq4)[5.])[1.#INF])[1.#INF]

What do you see instead?

Bremer Supports: Support tree: (Seq1 (Seq2 (Seq3 Seq4)[5.])[1073740250.])[1.#INF]

What version of the product are you using?

2602

On what operating system?

Windows

Please provide any additional information below.

On the simple test dataset (eq97.fas) with the same script, I get the expected output:

Bremer Supports: Support tree: (Seq1 (Seq2 (Seq3 Seq4)[3.])[1.#INF])[1.#INF]

Thanks for the report. This doesn't affect correctness. It is only a cosmetic issue, we should probably not even print infinity nor the maximum integer and leave the support value empty for trivial branches.

I agree, Andres - sounds good.

Mark

Turns out that removing the infinity is not really ... doable for now. I will partially fix the issue by getting rid of the big number and leaving the infinity, and will leave the elimination of infinity (which would require some more work), for version 4.0.1.