waldronlab / BugSigDBcuration

For documenting issues related to BugSigDB curation.
10 stars 7 forks source link

Integrating metagenomics with metabolomics for gut microbiota and metabolites profiling in acute pancreatitis #501

Closed SvetlanaUP closed 2 weeks ago

SvetlanaUP commented 1 month ago

Integrating metagenomics with metabolomics for gut microbiota and metabolites profiling in acute pancreatitis – Yan Jia – Scientific Reports https://www.nature.com/articles/s41598-024-72057-z

KateRasheed commented 1 month ago

I would like to claim this. Kate Rasheed

KateRasheed commented 1 month ago

Good day @SvetlanaUP This article is ready for review. Here is the link: https://bugsigdb.org/Study_1110

Signature 2 under Experiment 3 is not showing in the study. Here is the link for it: https://bugsigdb.org/Study_1110/Experiment_3/Signature_2

KateRasheed commented 1 month ago

Good day @SvetlanaUP . After I submitted my paper for review, the following edits were made:

  1. I removed all taxa in signatures(under experiment 2-4) with p-value threshold above 0.05.
  2. I added more taxa to signatures 2 (under experiment 2-4) where the taxa was increased abundance in the control group.

Thank you.

KateRasheed commented 1 month ago

Good morning @SvetlanaUP . Please my second contribution is yet to be reviewed.

SvetlanaUP commented 1 month ago

@KateRasheed It's on our to do list. Thank you for your patience! And, enjoy your weekend.

Folakunmi21 commented 3 weeks ago

Hello @KateRasheed. Thanks for your effort on this curation. Unfortunately, there were a lot of errors in it. You curated experiment 1 from fig 6 which is not curatable. A venn plot represents the relationship between different sets, but is not a statistical test. Alpha diversity measurements with p>0.05(i.e non-significant) should not be ignored. They should be curated as “unchanged”. The study should be curated as 3 experiments only; one experiment each for the lefse comparisons. -Experiment 1: control vs caerulein (signature source: fig 4C only) -Experiment 2: control vs caerulein+LPS (signature source: fig 4D only) -Experiment 3: control vs L-arginine (signature source: fig 4E only) Although ANCOM is a statistical test, Table S1 and fig S1 don’t show the groups that are increasing or decreasing. They should not be curated. Please, make these corrections so that I can mark the study as reviewed.

  1. All elements marked "Needs review" (none "Incomplete") (1 point): 1
  2. Correct study design (1 point): 1
  3. Entered all relevant experiments and no irrelevant experiments (1 point): 1
  4. Body site correctly identified (i.e. does not include multiple sites) (1 point): 1
  5. Condition entered according to contrast (correct EFO ontology) (1 point): 1
  6. Contrast groups correctly identified (1 point): 0.5
  7. Groups correctly labeled as 1 and 0 (1=cases, 0=controls) (1 point): 1
  8. Antibiotic exclusion correctly identified (1 point): 1
  9. Correctly identified sequencing details (2 points): 2
  10. Identified correct statistical test (1 point): 1
  11. Identified MHT correction (1 point): 1
  12. Correctly recorded matched on factors (1 point): 1
  13. Entered correct number of statistical tests per experiment (1 point): 1
  14. All diversity measures identified (1 point): 0
  15. Diversity results correctly entered as increased/decreased/unchanged (1 point): 0
  16. All signature sources correctly identified (-1 for each error) (2 points): 1
  17. Abundance direction correctly selected (1 point): 0
  18. Members of Signatures identified correctly (2 points): 0
  19. Correct use of NCBI taxonomy (2 points): 2

Total (maximum 23 points): 16.5

@SvetlanaUP Reviewed

KateRasheed commented 3 weeks ago

I'm really grateful for this review. I'll correct immediately.

For the supplementary table, I have taken note of the explanation as to why it should not be curated. I have a similar table in another study I'm collaborating on; I'll tag you to it on slack. Thank you once again @Folakunmi21

KateRasheed commented 3 weeks ago

Good day @Folakunmi21 . I have made corrections to this study. I have also tagged you to the table I need clarity on. Thank you so much.

SvetlanaUP commented 3 weeks ago

@Folakunmi21 please see corrections here, thanks!

Folakunmi21 commented 3 weeks ago

Thank you! @KateRasheed all changes reviewed

KateRasheed commented 3 weeks ago

Good day @SvetlanaUP . Please this is the issue Chloe gave clarity on.

SvetlanaUP commented 3 weeks ago

@KateRasheed please write down/document here the specific clarity.

KateRasheed commented 3 weeks ago

Okay. Chloe mentioned that the supplementary table is curatable. The log fold change shows the decreased/increased value for the case group. @SvetlanaUP

Folakunmi21 commented 3 weeks ago

Hi @KateRasheed I apologize for the oversight on the supplementary table. Please, curate it and I will adjust your score for point 3 in the marking guide. You combined the signature sources initially so I'll still have to deduct a mark for that (point 16 in the marking guide) but I'll give you 1 mark for it. Your combination of different signature sources affected the signatures you curated as well as their abundance directions so points 17 and 18 will have to remain that way. Sorry for the double work once again!

KateRasheed commented 3 weeks ago

It's fine. I'll get to work. Thank you so much.

KateRasheed commented 3 weeks ago

Good morning to you @Folakunmi21 . I have added all the signatures in the supplementary table. Thank you so much.

Folakunmi21 commented 3 weeks ago

Hi Kate @KateRasheed they don't seem to be reflecting yet... I can't see any signatures in the new experimnets you just added

KateRasheed commented 3 weeks ago

Yes yes. I noticed it as well. The signatures will show when they are entered manually.

For example: https://bugsigdb.org/Study_1110/Experiment_6/Signature_2

@Folakunmi21

KateRasheed commented 2 weeks ago

Good day @Folakunmi21 . All signatures are now showing on the study. I had to make edits (adding LEfSe statistical test and removing it from the experiments without signatures), because I observed that editing the experiments would make previous signatures show.

Thank you so much.

Folakunmi21 commented 2 weeks ago

Good job! @KateRasheed I've adjusted your points @SvetlanaUP all sorted✅