add-lancet-vardict-ngscheck-links-clin-harm

[jharenza]

Purpose

Briefly describe your changes here.
Add methods for:
-Lancet
-Vardict
-NGSCheckmate
-Clinical data harmonization (disease types, MB subtyping, OS)
-Add links to D3b github workflows

Issue

What GitHub issue does your pull request address?

Pull review checklist

The PR will be considered ready for review when all four items have been checked.

• All changes to text follow "one sentence per line" [Manubot instructions]
• All citations follow the Manubot citation instructions
• All changes or additions to tables follow the Manubot table instructions
• All figures follow the Manubot figure instructions

[cgreene]

A few thoughts - thanks for filling in these sections!

[jaclyn-taroni]

I had a few additional comments, thanks @jharenza!

[jaclyn-taroni]

@jharenza looks like an empty .Rhistory file snuck in with 6262e9d

[jharenza]

@jharenza looks like an empty .Rhistory file snuck in with 6262e9d

deleted

[jharenza]

@cansav09 - sorry I think I lost your comments when I committed. To answer your questions - I am going to work with @yuankunzhu and @migbro to make the BED files available. For WGS, BED regions were only used in Lancet (so it would not run forever). So we will get one BED file for WXS/Panel, which was used for all variant callers for that data type, then one BED for WGS Lancet. This could be used for TMB in the cases where the BEDs were used.

[cansavvy]

@cansav09 - sorry I think I lost your comments when I committed. To answer your questions - I am going to work with @yuankunzhu and @migbro to make the BED files available. For WGS, BED regions were only used in Lancet (so it would not run forever). So we will get one BED file for WXS/Panel, which was used for all variant callers for that data type, then one BED for WGS Lancet. This could be used for TMB in the cases where the BEDs were used.

Sounds great! I will continue using the BED file I have then as a placeholder and then once those are ready, switch that in.

[jharenza]

@cansav09 - sorry I think I lost your comments when I committed. To answer your questions - I am going to work with @yuankunzhu and @migbro to make the BED files available. For WGS, BED regions were only used in Lancet (so it would not run forever). So we will get one BED file for WXS/Panel, which was used for all variant callers for that data type, then one BED for WGS Lancet. This could be used for TMB in the cases where the BEDs were used.

Sounds great! I will continue using the BED file I have then as a placeholder and then once those are ready, switch that in.

OK - I am going to try to have these in the V5 release for tomorrow :)

[jaclyn-taroni]

I believe the only outstanding thing is this comment from @cgreene (#44 (comment)):

Are you looking at common variants across the population here? If so, I think that's the point to use polymorphism. Otherwise, I think variant is probably preferred. It wasn't clear to me that this was a point where the subject was focused on common variants. Maybe this means that SNP is still preferred but the sentence could be revised for clarity.

Provided @cgreene is satisfied with the disease type section

[jharenza]

I believe the only outstanding thing is this comment from @cgreene (#44 (comment)):

Are you looking at common variants across the population here? If so, I think that's the point to use polymorphism. Otherwise, I think variant is probably preferred. It wasn't clear to me that this was a point where the subject was focused on common variants. Maybe this means that SNP is still preferred but the sentence could be revised for clarity.

Provided @cgreene is satisfied with the disease type section

Yes, the way that many of the matching algorithms work are by looking at common, yet discriminatory SNPs within individuals (short of STR-typing).

[cgreene]

I am satisfied with disease type section.

It's still not entirely clear to me that you want polymorphism there. Are you only calling variants that are common in the population here?

[jharenza]

I am satisfied with disease type section.

It's still not entirely clear to me that you want polymorphism there. Are you only calling variants that are common in the population here?

Yes, we only use common SNPs here (similar to the days pre-STR for genotyping) - I will update the text.

Also, I will add some RNA QC.

[jharenza]

I am satisfied with disease type section.
It's still not entirely clear to me that you want polymorphism there. Are you only calling variants that are common in the population here?

Yes, we only use common SNPs here (similar to the days pre-STR for genotyping) - I will update the text.

Also, I will add some RNA QC.

@cgreene I have updated this section - do you want to check?

[cgreene]

👍