NGS tertiary analysis

Question

[n.awad]

I have a vcf file with different variant types, and I need to run tertiary analysis (variant classification, actionability, and annotation). Is there a way to get this information without searching for every variant in every database?

I think API is the best option but I don't know If the databases I am planning to use like (clinvar, COSMIC, and clingen) allow me to access this information via API or not!

Can you please provide me with guidance regarding this issue?

I appreciate your support,

Thanks,
Nour

[GenoMax]

API's for

clinvar - https://www.ncbi.nlm.nih.gov/clinvar/docs/maintenance_use/
COSMIC - https://clinicaltables.nlm.nih.gov/apidoc/cosmic/v3/doc.html
clingen - https://search.clinicalgenome.org/kb/downloads

Answer 1

[LauferVA]

nour - im trying to help so please dont be upset or offended...

the idea of getting access to a couple APIs and annotating some variants is anywhere from suboptimal (if a research question) to dangerous (if used for patient care). i dont know what you are planning to build or why, but id urge you to slow down and do some reading. let me explain.

at this point, every reputable database for sequence variant interpretation is going to have an API - finding APIs is not the problem. there are many tools, many databases, many software packages, many APIs ... if your approach is just "what are some APIs" you may finish quickly, but will the annotations mean anything? who knows - some are out of date, some may not fit your goals (clinical vs. something else?) etc.

what you need to do is the opposite. go slowly, tabulate the available options, then understand what each data source contains, how it is constructed, why it is used by people, and what makes it better or worse for your goals than the comparable tools.

depending on your goals, consensus recommendations on variant annotation may exist. we dont know if your data are clinical, bioinformatic, or what.. so its hard for us to guess what the right statement could be. AMP/ACMG could be a good fit; the 2015 guidelines are here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4544753/ since then many articles discussing it have been written and the guidelines themselves updated..

a search for "variant interpretation guidelines" returns: https://www.sciencedirect.com/science/article/pii/S0002929720303566 https://www.nature.com/articles/gim201842

but even here, just knowing which tools AMP/ACMG recommends isnt the best way to go. Its knowing WHY those are recommended.

[n.awad]

Hi Laufer, I really appreciate your response, it helps me a lot. your answer made me think more realistic about the results I would have, yeah it depends on my goal which is "clinical", I will take the analysis with extreme care as it will reflect human health.

Many thanks again for your support

[LauferVA]

:-) im happy to hear that. if clinical, i can take time to help if you want to.

for now, what do you think about this: https://clinicalgenome.org/working-groups/sequence-variant-interpretation/

[n.awad]

This is extremely supportive data, Thanks you so much

I will go through these links and I believe that I will have some questions or issues regarding interpretations, so If you have time to help me, It would be much appreciated.

[LauferVA]

look at this part of the website too (using PMS2 as example)

https://search.clinicalgenome.org/kb/genes/HGNC:9122

Answer 2

[bharata1803]

Hi, I think this software is the best for you: https://illumina.github.io/IlluminaConnectedAnnotationsDocumentation/ Illumina Connected Annotation accept VCF file and it will annotate using both transcript data and supplementary data. It has ClinVar and ClinGen supplementary database so you can see the relevant information for each variant from ClinVar and ClinGen. COSMIC is available but you have to purchase license or if you are academic user, you can get free by contacting Illumina.