search

vanheeringen-lab / ANANSE

Prediction of key transcription factors in cell fate determination using enhancer networks. See full ANANSE documentation for detailed installation instructions and usage examples.
http://anansepy.readthedocs.io
MIT License
77 stars 16 forks source link

Non-helpfull error region length #90

Open JGASmits opened 3 years ago

JGASmits commented 3 years ago

When running ananse binding with a region file with regions of only 1bp length, the error is not very descriptive:

2021-06-14 16:45:02 | DEBUG | Using default motif file 2021-06-14 16:45:44 | DEBUG | using motifs for 828 factors 2021-06-14 16:45:46 | INFO | Scanning regions for motifs. 2021-06-14 16:45:46,517 - INFO - reading table 2021-06-14 16:46:50,737 - INFO - creating score table (z-score, GC%) 2021-06-14 17:59:30,352 - INFO - done 2021-06-14 17:59:49,052 - INFO - creating dataframe 2021-06-14 18:05:56 | INFO | loading H3K27ac data 2021-06-14 18:21:11 | DEBUG | quantile normalization for H3K27ac 2021-06-14 18:21:13 | INFO | Loading models 2021-06-14 18:21:13 | INFO | 238 models found 2021-06-14 18:21:13 | INFO | Predicting TF activity 2021-06-14 18:22:07,013 - INFO - motif scanning (scores) 2021-06-14 18:22:07,014 - INFO - reading table 2021-06-14 18:23:17,086 - INFO - creating score table (z-score, GC%) 2021-06-14 18:28:41,333 - INFO - done 2021-06-14 18:29:03,692 - INFO - creating dataframe 2021-06-14 18:29:13,310 - INFO - Fitting BayesianRidge 0%| | 0/1 [00:00<?, ?it/s] 0%| | 0/1 [00:00<?, ?it/s] Traceback (most recent call last): File "/mbshome/ibonthuis/miniconda3/envs/ananse/bin/ananse", line 317, in args.func(args) File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/ananse/commands/binding.py", line 11, in binding predict_peaks( File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/ananse/peakpredictor.py", line 697, in predict_peaks p.predict_factor_activity(outfile) File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/ananse/peakpredictor.py", line 505, in predict_factor_activity activity[:] = minmax_scale(activity) File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/sklearn/utils/validation.py", line 63, in inner_f return f(*args, *kwargs) File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/sklearn/preprocessing/_data.py", line 545, in minmax_scale X = check_array(X, copy=False, ensure_2d=False, File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/sklearn/utils/validation.py", line 63, in inner_f return f(args, **kwargs) File "/mbshome/ibonthuis/miniconda3/envs/ananse/lib/python3.9/site-packages/sklearn/utils/validation.py", line 726, in check_array raise ValueError("Found array with %d sample(s) (shape=%s) while a" ValueError: Found array with 0 sample(s) (shape=(0,)) while a minimum of 1 is required.

I'm rerunning it now in order to see if adjusting region length fixes the issue. Ill comment once I'm sure the region lengths were the problem.

Ideal would be if with smaller region length ananse would resize the regions like some gimme motif tools. Alternatively an more insightfull error would be nice.

:)

JGASmits commented 3 years ago

It was indeed the region lenght that was the problem. :)