You may want to add an email address to your HN profile. Or have some method of contact listed.
I do something very similar in a research lab, and while it’s possible to make decent headway in this without much training, there are dragons all over the place.
For example, you didn’t mention which reference genome was used for the alignment/variant calling. Unless you use the right version for annotation, you’ll just get junk annotations that won’t make sense. I’ve only seen a couple of comments mention this.
If you don’t have the background, you might also need a crash course in human genetics, inheritance, molecular biology and variant functional prediction. You don’t need to become an expert, but you will need a working knowledge so that you know which variants to ignore. There really should be a small handful that would potentially make sense as a causal variant.
If the condition is sufficiently rare, you may not find clinical annotations, so be prepared to look a little deeper.
Comments
You may want to add an email address to your HN profile. Or have some method of contact listed.
I do something very similar in a research lab, and while it’s possible to make decent headway in this without much training, there are dragons all over the place.
For example, you didn’t mention which reference genome was used for the alignment/variant calling. Unless you use the right version for annotation, you’ll just get junk annotations that won’t make sense. I’ve only seen a couple of comments mention this.
If you don’t have the background, you might also need a crash course in human genetics, inheritance, molecular biology and variant functional prediction. You don’t need to become an expert, but you will need a working knowledge so that you know which variants to ignore. There really should be a small handful that would potentially make sense as a causal variant.
If the condition is sufficiently rare, you may not find clinical annotations, so be prepared to look a little deeper.
Best of luck.