failure to enter to the last step - Combining outputs

[liu3zhenlab]

"snps.txt" was produced but no log report about "Combining outputs". Everything else looks fine. I use conda installation in CentOS Linux 7. Appreciate your guide to solve the problem. Thanks.

Below is the error:

/.conda/envs/syri/lib/python3.5/site-packages/pandas/core/ops.py:1167: FutureWarning: elementwise comparison failed; returning scalar instead,
but in the future will perform elementwise comparison
result = method(y)
Traceback (most recent call last):
File "/software/syri/syri/bin/syri", line 255, in
getshv(args, coords, chrlink)
File "syri/pyxFiles/findshv.pyx", line 196, in syri.findshv.getshv
File "/.conda/envs/syri/lib/python3.5/site-packages/pandas/core/ops.py", line 1283, in wrapper
res = na_op(values, other)
File "/.conda/envs/syri/lib/python3.5/site-packages/pandas/core/ops.py", line 1169, in na_op
raise TypeError("invalid type comparison")
TypeError: invalid type comparison

[mnshgl0110]

Hi Sanzhen,

What version of pandas are you using? As of now I would guess that this is a pandas version issue. You could refer to the installation guide here to try and reinstall SyRI. You could also try creating a new python virtual environment and install syri there.

If the problem persists, then could you please run syri with the DEBUG mode and share the syri.log file? Also, it would be helpful to see the snps.txt file.

Best
Manish

[liu3zhenlab]

pandas version is 0.23.4. Is this the right version?

I skipped snp/indel calling using --nosnp but the final combination step had an error as well.

getTSV - ERROR - Error in finding parent for SV
Traceback (most recent call last):
File "/homes/liu3zhen/software/syri/syri/bin/syri", line 269, in
getTSV(args.dir, args.prefix, args.ref.name)
File "syri/pyxFiles/writeout.pyx", line 152, in syri.writeout.getTSV
AttributeError: 'Pandas' object has no attribute 'to_string'

[liu3zhenlab]

wc -l snps.txt 

23062458 snps.txt

head snps.txt

686 C . 18753100 20 686 26 1 1 1 1 1
706 A G 18753120 20 706 26 1 1 1 1 1
760 A . 18753173 54 760 26 1 1 1 1 1
866 . G 18753280 36 866 27 1 1 1 1 1
900 C . 156185066 1 900 28 1 1 1 1 7
901 G . 156185066 1 901 28 1 1 1 1 7
902 A . 18753315 1 902 28 1 1 1 1 1
902 A . 156185066 1 902 28 1 1 1 1 7
903 C . 18753315 1 903 28 1 1 1 1 1
938 C T 156185102 36 938 28 1 1 1 1 7

tail snps.txt 

159769419 G C 116433981 1 364 0 6 1 -1 9 9
159769437 T C 116433963 18 346 0 6 1 -1 9 9
159769556 A G 116433844 24 227 0 6 1 -1 9 9
159769580 . T 116433819 24 203 0 6 1 -1 9 9
159769617 T G 116433782 24 166 0 6 1 -1 9 9
159769641 C A 116433758 24 142 0 6 1 -1 9 9
159769676 T G 116433723 8 107 0 6 1 -1 9 9
159769684 A G 116433715 2 99 0 6 1 -1 9 9
159769686 . A 116433712 2 97 0 6 1 -1 9 9
159769722 A G 116433676 36 61 0 6 1 -1 9 9

[liu3zhenlab]

seems that chromosome ID in the snps.txt are not correct.

[mnshgl0110]

I think this could be happening because chromosome IDs are integers. Changing chromosome ids to 'chr1', 'chr2' etc might resolve this issue.
This would need to be done in original genome fasta files as well as in the alignments.
And you might need to rerun SyRI from beginning, to ensure that there are no inconsistencies.

[liu3zhenlab]

Thank Manish. I will try as you suggested.

[liu3zhenlab]

Hi Manish,
I used the following commands to modify sequence names and the delta file. After modifications, the SyRI ran successfully. Thank you very much for suggestions.

sed 's/^>\([0-9]\)/>chr\1/g'
sed 's/^>\([0-9]\{1,2\}\) \([0-9]\)/>chr\1 chr\2/g'