benchmark json creation

[emispano]

I was trying to use the evaluation tool and in the process of creating the preliminary files (when running raw_benchmark_parser.py for the first time) I encounter this problem:

READING /path/to/file/bpo_PSESM_type1.txt
READING /path/to/file/bpo_RAT_type1.txt
READING /path/to/file/bpo_DICDI_type1.txt
READING /path/to/file/bpo_SCHPO_type1.txt
READING /path/to/file/bpo_SULSO_type1.txt
READING /path/to/file/bpo_DROME_type1.txt
READING /path/to/file/bpo_HUMAN_type1.txt
READING /path/to/file/bpo_STRPN_type1.txt
READING /path/to/file/bpo_SALCH_type1.txt
READING /path/to/file/bpo_SALTY_type1.txt
READING /path/to/file/bpo_MYCGE_type1.txt
READING /path/to/file/bpo_DANRE_type1.txt
READING /path/to/file/bpo_PSEPK_type1.txt
READING /path/to/file/bpo_ARATH_type1.txt
READING /path/to/file/bpo_all_type1.txt
READING /path/to/file/bpo_XENLA_type1.txt
READING /path/to/file/bpo_eukarya_type1.txt
READING /path/to/file/bpo_prokarya_type1.txt
READING /path/to/file/bpo_METJA_type1.txt
READING /path/to/file/bpo_BACSU_type1.txt
READING /path/to/file/bpo_ECOLI_type1.txt
READING /path/to/file/bpo_HELPY_type1.txt
READING /path/to/file/bpo_MOUSE_type1.txt
READING /path/to/file/bpo_YEAST_type1.txt
READING /path/to/file/propagation_map_df_BPO.pkl
WRITING /path/to/file/BPO_MYCGE_243273_type_1_benchmark.json
WRITING /path/to/file/BPO_METJA_243232_type_1_benchmark.json
WRITING /path/to/file/BPO_ECOLI__type_1_benchmark.json
WRITING /path/to/file/BPO_BACSU__type_1_benchmark.json
WRITING /path/to/file/BPO_SALTY_99287_type_1_benchmark.json
WRITING /path/to/file/BPO_XENLA_8355_type_1_benchmark.json
WRITING /path/to/file/BPO_HELPY_85962_type_1_benchmark.json
WRITING /path/to/file/BPO_PSEPK_160488_type_1_benchmark.json
COULD NOT PROPAGATE bpo TERM GO:1905867
COULD NOT PROPAGATE bpo TERM GO:1905395
COULD NOT PROPAGATE bpo TERM GO:1905783
COULD NOT PROPAGATE bpo TERM GO:0099139
COULD NOT PROPAGATE bpo TERM GO:1905881
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:1905719
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:1905748
COULD NOT PROPAGATE bpo TERM GO:1905747
WRITING /path/to/file/BPO_eukarya__type_1_benchmark.json
COULD NOT PROPAGATE bpo TERM GO:0061844
COULD NOT PROPAGATE bpo TERM GO:0061844
WRITING /path/to/file/BPO_MOUSE_10090_type_1_benchmark.json
WRITING /path/to/file/BPO_YEAST_559292_type_1_benchmark.json

(the output is similar for MFO and CCO)

What I'm asking is:

• why are there GO terms present in groundtruth but not in the given obo file?
• why do i get the json files for only some of the species that were in the starting list? Does this have anything to do with some GOs not being properly propagating?

I am using the example data in https://iastate.app.box.com/v/cafa-evaluation-example-data for the obo file, for the groundtruth (leafonly_.txt) and for the target lists (__type.txt)