Clinical data release 2

[fpbarthel]

Clinical related to-do's for data release 2

• [x] Some of the barcodes have incorrect sample types. @moskalika made a list of the ones we've currently identified. This is not a priority to fix, but will keep track of inconsistencies here as to address these later. Each item is as follows: sample_barcode | surgery_number | sample_number
• GLSS-MD-0087-R1 | 4 | R1
• GLSS-MD-0087-TP | 2 | TP
• GLSS-MD-0091-R1 | 3 | R1
• GLSS-MD-0091-TP | 2 | TP
• GLSS-MD-0093-R1 | 3 | R1
• GLSS-MD-0093-TP | 2 | TP
• GLSS-MD-0096-R1 | 3 | R1
• GLSS-MD-0089-R1 | 3 | R1
• GLSS-MD-0089-TP | 2 | TP
• GLSS-HF-2869-TP | 3 | TP
• GLSS-HF-3081-TP | 2 | TP
• GLSS-MD-0113-R1 | 1 | R1
• GLSS-MD-0113-R4 | 2 | R2
• GLSS-MD-LP03-R1 | 3 | R1
• GLSS-MD-LP03-TP | 2 | TP

Sample mixups

• [x] Consolidate GLSS-MD-0027 and GLSS-MD-0095 as these were found to be the same patient as per https://github.com/TheJacksonLaboratory/GLASS/issues/102

[fpbarthel]

@moskalika @Kcjohnson please comment on any clinical-related to-do's here

[moskalika]

• [x] Consolidate GLSS-MD-0027 and GLSS-MD-0095

• [x] Remove following case_barcodes from clinical.surgeries and clinical.cases tables (looks like I don't have permission to DELETE FROM):

• [x] 'GLSS-MD-0095'

@fpbarthel: leaving the following cases in the dataset despite marking the associated sequencing data as deprecated, in case we ever get high quality data for these.

GLSS-MD-0019 GLSS-MD-0084 GLSS-MD-0085 GLSS-MD-0137 GLSS-MD-LP03 GLSS-MD-LP07 GLSS-MD-LP08

[fpbarthel]

For GLSS-MD-0027 and GLSS-MD-0095, which ID are we going to toss out?

[moskalika]

'GLSS-MD-0095'

[fpbarthel]

@moskalika can you check this to-do list https://github.com/TheJacksonLaboratory/GLASS/issues/123 if i forget anything or made a mistake

[fpbarthel]

@moskalika I'll leave the non-duplicated cases in the database for now and have noted that this data is deprecated in the analysis.files table. Should we ever receive quality data for these samples we can easily add it. We should take care to remove these cases from the supplement for the paper though.

[fpbarthel]

Since fixing all the barcodes is a bit too much work, we'll accept the fact that some of the sample_type designations are incorrect. Users can check the surgeries table for correct labels.

[Kcjohnson]

We caught an error with regards to tmz_treament. The clinical.surgeries table previously did not list GLSS-MD-0036-TP sample as having received TMZ, despite this sample displaying hypermutation at recurrence. I made the following database update on 2019.04.23. It is our intent to have another release with all the analyses that make their way into the final manuscript.

rs = dbSendStatement(con, "UPDATE clinical.surgeries SET treatment_tmz = '1' , treatment_tmz_cycles = '10' WHERE sample_barcode = 'GLSS-MD-0036-TP'")