Hello,
I have a test dataset containing five normal and five tumor samples. I did majiq build for 10 samples individually first.
majiq --license {} build {} -c {} -o {} -j {} --min-experiments {} \
--simplify {} --min-denovo 3
Then, I executed the following command:
majiq --license $license_dir heterogen -j 4 -o $output_dir/heterogen \
-n tumor normal \
-grp1 $output_dir/tumor/majiq/*/*.majiq \
-grp2 $output_dir/normal/majiq/*/*.majiq
It can parse the required data. But then returns error message as below:
```
2024-08-01 11:27:17,483 (PID:1902070) - INFO - Group normal: 34480 LSVs
2024-08-01 11:27:17,500 (PID:1902070) - INFO - Number quantifiable LSVs: 25408
2024-08-01 11:27:17,608 (PID:1902070) - INFO - Sampling from PSI
2024-08-01 11:27:17,627 (PID:1902070) - INFO - Group tumor performing at least 501 psi samples per experiment to bound visualization error
2024-08-01 11:27:17,627 (PID:1902070) - INFO - (group 1/2, experiment 1/5) Group tumor sampling PSI from '/data/TCGA/output/tumor/majiq/295c896a-e9c1-467b-ae56-15b18aad97e8/ef697db7-3b14-4999-9e9b-85c37718f615.rna_seq.genomic.gdc_realn.majiq'
Traceback (most recent call last):
File "rna_majiq/src/io.pyx", line 298, in rna_majiq.src.io.get_coverage_mat_lsv
IndexError: index 5 is out of bounds for axis 0 with size 5
Exception ignored in: 'rna_majiq.src.io.get_coverage_mat_lsv'
Traceback (most recent call last):
File "rna_majiq/src/io.pyx", line 298, in rna_majiq.src.io.get_coverage_mat_lsv
IndexError: index 5 is out of bounds for axis 0 with size 5
Segmentation fault (core dumped)
```
May I ask:
1. Does the [experiment] group name in configuration .ini, which was used at build step, affect the behavior at quantification step? Should the group name in .ini be exactly the same as the arguments used in quantification step?
2. Would you mind helping me if there is any other possible reason that may cause this error?
Thank you very much!
Best,
Yihao
2.