zstephens
commented
7 years ago Hey there,
That error is indicating that on an attempt to modify the reference with an indel the reference nucleotide to be altered was not as expected. From the command you provided I don’t see how any mutations could’ve even been inserted (-M 0 means no random mutations, was there a -v parameter that was omitted?).
I see that the variant that was inserted (my guess is from -v) is a multi-SNP (if that terminology is standard) where 2 bases are subbed out for 2 others. At the moment I believe NEAT is only set up to accept variants in the following form:
SNP: (position=X, ref=A, alt=B) INS: (position=X, ref=A, alt=ABCDE…) DEL: (position=X, ref=ABCDE…, alt=A)
So the unexpected format caused the error. Acceptable indels are either one-to-many or many-to-one, as specified above. Multi-SNPS (e.g. 2-to-2, in your example) and similarly complex insertion-deletion combos (e.g. ref=ACGT, alt=AA) are not yet supported. Here are my ideas for how to move forward on this:
1) Ideally I’d like to add full support for arbitrarily complex events (such that they appear in the output vcf just as they do in the input)
2) Slightly-less ideal would be to break them up into multiple sub-events. E.g. if the input VCF contained:
14487, TCC —> TTT
The output VCF would contain:
14488, C —> T 14489, C —> T
3.) At a bare minimum the events could be ignored, and a warning could be printed to console.
If option 2 is acceptable I’ll take a crack at implementing that later today. If not then I’ll implement option 3 and eventually get around to fully supporting the events a la option 1.
Cheers, Zach
On Oct 12, 2016, at 12:09 PM, Stephen J Bush notifications@github.com<mailto:notifications@github.com> wrote:
At higher coverage paramters (e.g., -c 500), I am getting the following quite often:
Error: Something went wrong! (14487, 'TCC', 'TTT') TTC
from this command
for i in $(seq 1 10) do ( python /neat-genreads/genReads.py \ -r ${REFERENCE} \ -R 126 \ -o "${prefix}" \ --pe 300 30 \ -t "${REGIONS}" \ -c 500 \ -M 0 \ --bam --vcf \ --job $i 10 ) & done
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://urldefense.proofpoint.com/v2/url?u=https-3Agithub.com_zstephens_neat-2Dgenreads_issues_7&d=DQMCaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=iSEBNMfCHfS5bLIS7AqI6wkxgRNv8-2dMwRF43ddddU&m=b2ZBMwFEiVNv3WZwMvc6CIvKfmzaihavV8Uqg3KomQM&s=PSPgbVVxCY9AxrQHE71m78zrvLY8hlYSSI3aFDcjqRg&e=, or mute the threadhttps://urldefense.proofpoint.com/v2/url?u=https-3Agithub.com_notifications_unsubscribe-2Dauth_AIHT1fQPmM8ojMOar-5FbmregigGXmkvWDks5qzRQ9gaJpZM4KU-5Ffq&d=DQMCaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=iSEBNMfCHfS5bLIS7AqI6wkxgRNv8-2dMwRF43ddddU&m=b2ZBMwFEiVNv3WZwMvc6CIvKfmzaihavV8Uqg3KomQM&s=ORsvAM1Us5d-KAxSQxzOkk5q0YeGE9liSn3HwuQol2c&e=.
muppetjones
At higher coverage paramters (e.g.,
-c 500), I am getting the following quite often:from this command
I am unsure what the exact error is, but as the script exists when this occurs, the resulting files are incomplete.