Minor improvements for custom database annotation #67
Conversation
Blast database throw error for no alias or index file found for nucleotide database, when using blastn. Thus, make blast database before execute blastn to custom database on genome.
Whitespace occurs when generated links for custom database report in Shiny markdown. Specify seperate string to emtpy string to fix this.
|
Hi @lam-c, I will take a look at the changes later today to review the changes. I'll also let the example datasets running so we can make sure all works. If all goes good, we can may have all merged whithin this week. 🚀 👍🏼 |
|
Execution was successful for example dataset, now I just need to assess the changes on the shiny app. |
|
I am not sure about this comment:
Is it possible for you to send the logs for the report executions? Also, have you tested if your changes fixed this? I guess it may be just an adjusment on the 'loop' on the custom annotation results when running the REPORTS module. But, to be sure, would be nice to take a look at the loggings, and files, erros, etc. 😄 |
|
I think I just ran into the error you mentioned. The error is basically that, even though the summary files are not empty, the tables in the reports are empty, right? I will take a look on that. Can you open an issue for that? I will link to this PR. |
Exactly! I've opened an issue as you advised just now. The commond logs show that everything goes fine, I guess may be some bugs in Sorry for not replying sooner. MUCH lots of THANKS for prompt replies and active debugging! |
|
Hi @lam-c, Then, I will start to work on the issue you've opened which is not related to the workflow itself but to the shiny app which means I have to debug the shiny app code and this, should be tackled in another PR. Many thanks for the contributions. 😄 |
* update hmmer to working version with PGAP * update hmmer to working version with PGAP (#59) (#60) * Adding option for using `bakta` (#58) * first bakta addition test * Fixing error when phigaro output was empty (#55) (#56) * adding checkIfExists rules to input loading * Update build.sh * fixed report if else statement * updated version and changelog * fix example samplesheet url * Create nf-core-bacannot-compare_logo_dark.png * Update phigaro.nf * Update CHANGELOG.md * Update CHANGELOG.md * Update .gitignore * Update bacannot.nf * workflow runs from top to bottom * workflow correctly working * changing version manifest * explaining how to use bakta * adding variables to generic annotation * adding victors backup db * report_general understands bakta * report_resistance understands bakta * update hmmer to working version with PGAP * update hmmer to working version with PGAP (#59) * update hmmer to working version with PGAP (#59) (#60) * Update resfinder2gff.py * Update addBedtoolsIntersect.R * Update addBedtoolsIntersect.R * Update addBedtoolsIntersect.R * Update addBedtoolsIntersect.R * Update docker.config * download db if not available * Update merge_annotations.nf * Update merge_annotations.nf * update process label * Update merge_annotations.nf * fix docker file ownerships * added retry label for bakta * update changelog * changing to new version * add falmeida-py scripts to docker * Create generate_dockers.sh * Minor improvements for custom database annotation (#67) * Add step making Blast DB when annotating custom db Blast database throw error for no alias or index file found for nucleotide database, when using blastn. Thus, make blast database before execute blastn to custom database on genome. * Remove space when generate link in Shiny markdown Whitespace occurs when generated links for custom database report in Shiny markdown. Specify seperate string to emtpy string to fix this. * fix: script was calling an out of bounds value * handle exceptions and missing values * increment test profile * also build for proteins * Use correct version of customdb_gffs in REPORTS module Co-authored-by: fmalmeida <[email protected]> * fixed docker images * resfinder now uses conda env * rgi does not have --exclude_nudge anymore * add circos conf files * fix indentation * add circos sub-workflow calling * add circos tool * added karyotype and gc_skew modules * fix indentation * modules now have a channel suitable for summary generation * include summary generation module * update docker image * add aro index tsv * add amrfinder2tsv module * add bacannot json summary generation * update docker images * add new modules * added binaries for circos plot * update docker files * add label to module * finalise modules preparing data for circos * include whole circos 'pathway' in workflow * add circos module * circos plot currently functional * re-organised circos outputs * ameliorate default version of circos plot * fixed docker image * fixed custom annotation problem in server * added circos plot with easy_circos * circos plot properly being generated * update changelog * update readme * update documentation * fix tool name * fix dir * update indentation * fixed schema * add pipefail * fix docker image Co-authored-by: Cy <[email protected]>
Some bugs happened when using custom databases for annotation, so I simply modified codes to solve them. Changes are roughly about:
blastnfor annotating custom database, to resolve errors lacking index file for nucleotide db;There still exist bugs in custom database reports. Some of the report html file are empty when assigning multiple custom databases, while the annotation files (
.summary.txt,.gff, both not empty) successfully completed. I'm not sure if this problem are caused by modification above.Thank you soooo much for your patience, and this awesome tool!