updates for metadata and manuscript finalization #2

gregcaporaso · 2024-10-08T16:09:39Z

No description provided.

if `export_legend` is enabled, present the legend in the resulting Visualization. Note that there was a minor bug in the original version of the Visualizer - since the conditional was being evaluated after `export_legend` was cast to a string, it always evaluated to `True`.

gregcaporaso · 2024-10-08T16:12:36Z

@lizgehret, this should be updated for the new metadata. Here's the bucket 3/4 plot that you create in your example command - does this align with your previous version?

Figure:

Legend:

lizgehret · 2024-10-08T16:41:39Z

Hey @gregcaporaso, I think there's a little confusion around the export_legend functionality - it should only be included in the main plot if export_legend is set to False. The purpose of this is to allow for Jeff to combine multiple bucket figures into a faceted view in biorender, with the legend exported as a separate figure that he can add in where it stylistically makes sense. The expected behavior is as follows:

if export_legend == False:
Only the pcoa_plot.png should be included in the data directory, and it should include the legend in the main figure.
if export_legend == True:
The pcoa_plot.png and legend.png should both be included in the data directory, and the legend should not be included in the main figure.

I just tested this again and confirmed it's working as expected. Do you think it would be helpful to change the parameter name to prevent any confusion on what the expected behavior is?

lizgehret · 2024-10-08T16:46:26Z

Re: the figure looking the same - it looks very similar, but not exactly identical. Here's what my version looks like:

I'm just seeing a few differences in the bulking material samples, which we may want to look at together during the composting meeting (just to make sure none of the Metadata changes unintentionally changed the identification of any particular samples).

lizgehret

i took a closer look at the export_legend stuff we were discussing @gregcaporaso - here are more detailed comments.

gut_to_soil_manuscript_figures/_methods.py

lizgehret · 2024-10-08T21:40:24Z

gut_to_soil_manuscript_figures/scripts/plot_pcoa_2d.py

@@ -62,8 +63,7 @@ def _bucket_util(highlighted_buckets, md, ord_2d):
        # HE
        bucket_ids_HE_week0 = \
            md[(md['Bucket'] == bucket) &
-               (md['SampleType2'] == 'Self Sample') &
-               (md['Week'] == 0.0)].index.values
+               (md['SampleType'] == 'Human Excrement')].index.values


It looks like the (md['Week'] == 0.0)].index.values got removed. Was this intentional?

lizgehret · 2024-10-08T21:41:08Z

gut_to_soil_manuscript_figures/scripts/plot_pcoa_2d.py

@@ -73,17 +73,16 @@ def _bucket_util(highlighted_buckets, md, ord_2d):
        # bulking
        bucket_ids_bulk_week0 = \
            md[(md['Bucket'] == bucket) &
-               (md['SampleType2'] == 'Bulking Material') &
-               (md['Week'] == 0.0)].index.values
+               (md['SampleType'] == 'Bulking Material')].index.values


Same comment as above, it looks like (md['Week'] == 0.0)].index.values was removed here as well.

lizgehret · 2024-10-08T21:42:39Z

gut_to_soil_manuscript_figures/scripts/plot_pcoa_2d.py

@@ -233,8 +235,7 @@ def plot_pcoa_2d(metadata_fp, ordination_fp, measure,
        if not highlighted_buckets:
            # HE wk 0 mean
            HE_week0 = \
-                md[(md['SampleType2'] == 'Self Sample') &
-                   (md['Week'] == 0.0)].index.values
+                md[(md['SampleType'] == 'Human Excrement')].index.values


Same comment re: (md['Week'] == 0.0)].index.values

lizgehret · 2024-10-08T21:42:45Z

gut_to_soil_manuscript_figures/scripts/plot_pcoa_2d.py

@@ -245,8 +246,7 @@ def plot_pcoa_2d(metadata_fp, ordination_fp, measure,

            # bulk wk 0 mean
            bulk_week0 = \
-                md[(md['SampleType2'] == 'Bulking Material') &
-                   (md['Week'] == 0.0)].index.values
+                md[(md['SampleType'] == 'Bulking Material')].index.values


Same comment re: (md['Week'] == 0.0)].index.values

Co-authored-by: Liz Gehret <[email protected]>

gregcaporaso · 2024-10-09T14:16:19Z

@lizgehret, thanks for catching the Week 0.0 issue! I think I fixed it now - we replaced those with empty cells, so checks for week == 0.0 are now checks for Composting Time Point is np.nan.

Here's the broken version I had yesterday for bucket 7:

Here's today's fixed version:

And here is one I'm pulling from Basecamp which I believe is the target:

Note that in our metadata clean-up process, a few of the FLWC samples were removed so that is almost certainly the source of minor differences in the ordination.

lizgehret · 2024-10-10T00:36:46Z

@gregcaporaso this does look better - the only thing I'm wondering about is the number of FLWC samples that are in your updated version (in comparison to the version from BC). It appears there's a higher number of that sample type than there was before - is that expected? We may want to do a quick tally of the total number we're expecting and visually see if that matches up with the updated plot.

gregcaporaso · 2024-10-17T15:29:05Z

Replaced by #4

gregcaporaso added 2 commits October 8, 2024 08:53

updates to reflect refactored metadata

b9da739

gregcaporaso requested a review from lizgehret October 8, 2024 16:09

lizgehret requested changes Oct 8, 2024

View reviewed changes

gregcaporaso and others added 5 commits October 9, 2024 06:46

Update gut_to_soil_manuscript_figures/_methods.py

0331476

Co-authored-by: Liz Gehret <[email protected]>

Update gut_to_soil_manuscript_figures/_methods.py

04f45c9

Co-authored-by: Liz Gehret <[email protected]>

Update gut_to_soil_manuscript_figures/_methods.py

fffaaec

Co-authored-by: Liz Gehret <[email protected]>

Update gut_to_soil_manuscript_figures/_methods.py

aabe60f

Co-authored-by: Liz Gehret <[email protected]>

legend_fp was referenced before assignment

ac3f992

gregcaporaso force-pushed the metadata-updates branch from dc7b6a6 to ac3f992 Compare October 9, 2024 14:03

address issue with missing Week 0.0 checks

c917fee

gregcaporaso closed this Oct 17, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

updates for metadata and manuscript finalization #2

updates for metadata and manuscript finalization #2

gregcaporaso commented Oct 8, 2024

gregcaporaso commented Oct 8, 2024

lizgehret commented Oct 8, 2024

lizgehret commented Oct 8, 2024 •

edited

Loading

lizgehret left a comment

lizgehret Oct 8, 2024

lizgehret Oct 8, 2024

lizgehret Oct 8, 2024

lizgehret Oct 8, 2024

gregcaporaso commented Oct 9, 2024

lizgehret commented Oct 10, 2024

gregcaporaso commented Oct 17, 2024

updates for metadata and manuscript finalization #2

updates for metadata and manuscript finalization #2

Conversation

gregcaporaso commented Oct 8, 2024

gregcaporaso commented Oct 8, 2024

lizgehret commented Oct 8, 2024

lizgehret commented Oct 8, 2024 • edited Loading

lizgehret left a comment

Choose a reason for hiding this comment

lizgehret Oct 8, 2024

Choose a reason for hiding this comment

lizgehret Oct 8, 2024

Choose a reason for hiding this comment

lizgehret Oct 8, 2024

Choose a reason for hiding this comment

lizgehret Oct 8, 2024

Choose a reason for hiding this comment

gregcaporaso commented Oct 9, 2024

lizgehret commented Oct 10, 2024

gregcaporaso commented Oct 17, 2024

lizgehret commented Oct 8, 2024 •

edited

Loading