ValueError: basic_string::substr in MinHash.add_sequence #502
Comments
|
When run locally, I get a similar error: ✘ Wed 20 Jun - 21:05 ~/kmer-hashing/sourmash/maca/facs_v4_1000cell_scaled_trim_comparison origin ☊ master 3☀ 6●
ubuntu@olgabot-reflow sourmash info
sourmash version 2.0.0a7
- loaded from path: /home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash
✘ Wed 20 Jun - 21:07 ~/kmer-hashing/sourmash/maca/facs_v4_1000cell_scaled_trim_comparison origin ☊ master 3☀ 6●
ubuntu@olgabot-reflow sourmash compute --track-abundance --protein --dna --scaled 100 --ksizes 21,33,51 --merge 'G14-MAA001857-3_38_F-1-1|tissue:Pancreas|subtissue:Endocrine|cell_ontology_class:pancreatic_A_cell|free_annotation:pancreatic_A_cell' --output G14-MAA001857-3_38_F-1-1.signature G14-MAA001857-3_38_F-1-1.trimmed
setting num_hashes to 0 because --scaled is set
computing signatures for files: G14-MAA001857-3_38_F-1-1.trimmed
Computing signature for ksizes: [21, 33, 51]
Computing both DNA and protein signatures.
Computing a total of 6 signature(s).
Tracking abundance of input k-mers.
... reading sequences from G14-MAA001857-3_38_F-1-1.trimmed
... G14-MAA001857-3_38_F-1-1.trimmed 900000Traceback (most recent call last):
File "/home/ubuntu/anaconda/bin/sourmash", line 6, in <module>
sys.exit(sourmash.__main__.main())
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/__main__.py", line 76, in main
cmd(sys.argv[2:])
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/commands.py", line 287, in compute
args.input_is_protein, args.check_sequence)
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/commands.py", line 197, in add_seq
E.add_sequence(seq, not check_sequence)
File "sourmash/_minhash.pyx", line 178, in sourmash._minhash.MinHash.add_sequence
ValueError: basic_string::substrIt seems interesting that with either file, the error happened after viewing kmer # |
|
Hmm it seems that something is off with checking the sequence because adding Wed 20 Jun - 21:19 ~/kmer-hashing/sourmash/maca/facs_v4_1000cell_scaled_trim_comparison origin ☊ master 3☀ 6●
ubuntu@olgabot-reflow sourmash compute --track-abundance --protein --dna --scaled 100 --ksizes 21,33,51 --merge 'G14-MAA001857-3_38_F-1-1|tissue:Pancreas|subtissue:Endocrine|cell_ontology_class:pancreatic_A_cell|free_annotation:pancreatic_A_cell' --output G14-MAA001857-3_38_F-1-1.signature --check-sequence G14-MAA001857-3_38_F-1-1.trimmed
setting num_hashes to 0 because --scaled is set
computing signatures for files: G14-MAA001857-3_38_F-1-1.trimmed
Computing signature for ksizes: [21, 33, 51]
Computing both DNA and protein signatures.
Computing a total of 6 signature(s).
Tracking abundance of input k-mers.
... reading sequences from G14-MAA001857-3_38_F-1-1.trimmed
Traceback (most recent call last):
File "/home/ubuntu/anaconda/bin/sourmash", line 6, in <module>
sys.exit(sourmash.__main__.main())
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/__main__.py", line 76, in main
cmd(sys.argv[2:])
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/commands.py", line 287, in compute
args.input_is_protein, args.check_sequence)
File "/home/ubuntu/anaconda/lib/python3.6/site-packages/sourmash/commands.py", line 197, in add_seq
E.add_sequence(seq, not check_sequence)
File "sourmash/_minhash.pyx", line 178, in sourmash._minhash.MinHash.add_sequence
ValueError: invalid DNA character in input: C |
|
For all of these, here is the input, trimmed fastq file:
|
|
well that just seems bad. we did fix lowercase problems in #480 but this
doesn't seem like the problem here.
|
|
On Wed, Jun 20, 2018 at 03:40:54PM -0700, Olga Botvinnik wrote:
For all of these, here is the input, trimmed fastq file: `s3://olgabot-maca/temp/G14-MAA001857-3_38_F-1-1.trimmed` (716 MB)
I don't appear to have access to this file; could you drop me the public
download link? thx!
|
|
sorry about that! The S3 prefix is public now. Here is a HTTP download link too, just in case. |
|
It looks like the problem is with this sequence: which has an invalid character (an 'N') in it. (The error message is reporting the wrong character, and I'll fix it when I track it down.) |
5 tasks
|
Fixed in #503. |
|
Ah okay great! For the offending Should the order be: Option A
Option B
|
|
I would recommend turning off --check-sequence, which will simply ignore
k-mers containing unknown characters.
|
|
Got it. I'm still getting the Here's the install command: And here's the |
|
I split the file into 10k seq pieces (40k lines) and After hunting down the error and |
|
Note: this only happens with |
|
Any suggestions? This problem still only occurs for me when I'm fine computing only DNA signatures for now but I'd like to do protein in the future, too. |
2 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello! I'm using the container
docker pull quay.io/biocontainers/sourmash:2.0.0a7--py27_0to run sourmash compute directly ontrim-low-abund.pyreads and am getting this strange error:If I use the
masterbranch and build with this dockerfile, then I get a different error but on the same line (178 insourmash/_minhash.pyx):For reference, here is the trimmed fastq file:
s3://olgabot-maca/temp/A10_B000419_S34.trimmedDo you know what may be happening?
The text was updated successfully, but these errors were encountered: