AG-1395: Bump harmonized targets source version to pick up delimiter fix

[JessterB]

No description provided.

[jaclynbeck-sage]

Actually I take back my approval until CI has re-run. I see this may have caused an issue with gene_info but want to let CI run again to make sure.

[jaclynbeck-sage]

Yeah ok, it looks like something about the new file did something bad to gene_info processing. What is different about the new file?

[JessterB]

Yeah ok, it looks like something about the new file did something bad to gene_info processing. What is different about the new file?

@jaclynbeck-sage I converted some semi-colon delimiters to commas, and forgot to escape them. I'll update the source and verify locally before I push up a new commit...sorry, I should have tested this locally first.

[jaclynbeck-sage]

Thanks. It looks like something is... off with this file (as well as its previous versions 48 and 47 which is as far back as I looked). The raw file has 1362 lines in it, but if you read it in as a CSV either programmatically or with Excel it only has 1162 lines (versions 48-50). For v47 it as 1334 lines / 1134 lines, so a 200-line discrepancy in both versions.

Update: It looks like starting line 320, The Emory data has line breaks in the Data_used_to_support_target_selection column, which are escaped, so it does get read in successfully as one field. I guess that's ok? The raw file looks like:

320  ...,"Discovery quantitative proteomics of FrCx 
321    WPCNA of multiple and consensus cohorts
322    ANOVA",...

and the read-in string looks like:
"Discovery quantitative proteomics of FrCx \n WPCNA of multiple and consensus cohorts\n ANOVA"

Is this something we're aware of/handle on the front end for display?

[JessterB]

Thanks. It looks like something is... off with this file (as well as its previous versions 48 and 47 which is as far back as I looked). The raw file has 1362 lines in it, but if you read it in as a CSV either programmatically or with Excel it only has 1162 lines (versions 48-50). For v47 it as 1334 lines / 1134 lines, so a 200-line discrepancy in both versions.

Update: It looks like starting line 320, The Emory data has line breaks in the Data_used_to_support_target_selection column, which are escaped, so it does get read in successfully as one field. I guess that's ok? The raw file looks like:

320  ...,"Discovery quantitative proteomics of FrCx 
321    WPCNA of multiple and consensus cohorts
322    ANOVA",...

and the read-in string looks like: "Discovery quantitative proteomics of FrCx \n WPCNA of multiple and consensus cohorts\n ANOVA"

Is this something we're aware of/handle on the front end for display?

@jaclynbeck-sage This seems to look ok in Agora in prod today (for the 2 genes I checked, RPH3A & STX1A): mv68 -> gene_info v56 -> harmonized targets v48

...but I could remove those /n since they are confusing and useless.

[jaclynbeck-sage]

...but I could remove those /n since they are confusing and useless.

Maybe? I'm guessing the line breaks are to indicate these are 3 separate items, not one single discovery method, so the way it's displaying in Agora right now isn't right because it looks like it's all mashed together in one sentence. I think probably the \n should be replaced with commas but also there's extra spaces around the \n that need to be dealt with in order for commas to look normal.

[sonarcloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[JessterB]

@jaclynbeck-sage Those newlines go back to v1 of the source file from 2018 and no one has complained yet, but I went ahead and fixed it (in v51) by replacing the newlines with commas, and cleaning up extra spaces.

[jaclynbeck-sage]

Looks good, thank you!!