[BENCHMARK] Run parameter benchmarks with different input data and update them. #207
Conversation
Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
[TEST] Update tests Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
Signed-off-by: Lydia Buntrock <lydia.buntrock@fu-berlin.de>
Codecov Report
@@ Coverage Diff @@
## master #207 +/- ##
=======================================
Coverage 98.35% 98.35%
=======================================
Files 18 18
Lines 850 850
=======================================
Hits 836 836
Misses 14 14 Continue to review full report at Codecov.
|
iGenVar only - new Plots:
|
| min_qual=list(range(config["quality_ranges"]["iGenVar"]["from"], | ||
| config["quality_ranges"]["iGenVar"]["to"], | ||
| config["quality_ranges"]["iGenVar"]["step"]))) |
There was a problem hiding this comment.
This is copied for in each input – is there a way to pre-define this?
| dataset=["Illumina_Paired_End", "Illumina_Mate_Pair", "MtSinai_PacBio", "PacBio_CCS", "10X_Genomics"], | ||
| parameter_name="partition_max_distance"), | ||
| expand("results/parameter_benchmarks/{dataset}/plots/{parameter_name}.results.all.png", | ||
| dataset=["Illumina_Paired_End", "Illumina_Mate_Pair", "MtSinai_PacBio", "PacBio_CCS", "10X_Genomics"], |
There was a problem hiding this comment.
the dataset is also always the same
There was a problem hiding this comment.
Good point, this appears in every benchmark workflow. So I think I will try to make it shorter in another PR.
With exactly this PR. So far I have tested iGenvar only on one dataset regarding the parameters, now I have added others. And then I changed only one parameter at a time and saw how our result behaves. With some back and forth I have now ended up with the dafault values and plots created here. On the plots you can see that larger or smaller values would not give any improvement. |
min_var_length = 30 (remains)
max_var_length = 10.000 (changed from 100.000)
max_tol_inserted_length = 50 (remains)
max_tol_deleted_length = 50 (remains)
max_overlap = 50 (changed from 10)
partition_max_distance = 50 (changed from 1.000)
hierarchical_clustering_cutoff = 0.3 (changed from 0.5)