search

greenelab / deep-review

A collaboratively written review paper on deep learning, genomics, and precision medicine
https://greenelab.github.io/deep-review/
Other
1.24k stars 270 forks source link

The precision-oncology illusion #105

Open cgreene opened 7 years ago

cgreene commented 7 years ago

http://doi.org/10.1038/537S63a

This perspective paper is relevant for our discussion of how we categorize disease. I think that these difficulties are something that we'll need to recognize in our paper. One might expect the area of cancer biology, where we have focused numerous sequencing efforts, to provide the strongest argument. But the evidence in favor of a precision oncology strategic inflection point remains weak, even in this area.

The foundation of this perspective is provided largely by: http://dx.doi.org/10.1016/j.mayocp.2015.08.017

agitter commented 7 years ago

I'm not up to date on the latest clinical precision oncology strategies, but matching targeted therapies to driver gene mutations is more of a starting point than an endpoint. After a quick read, I see this as motivation for methods like #6 (categorization beyond mutation status), #11 (and many others looking at the impact of non-coding variants), and #45 (and the related virtual screening methods at seek to provide better treatments in the very long term).

cgreene commented 7 years ago

@agitter : Agree - complexity of biology & robustness of living systems is such that we should not expect driver gene mutations -> targeted therapies to be a strategy that works in the general case. However, I've had this conversation with people a number of times, and it still remains a view that is held. I'm planning to include this to motivate the need for improved phenotyping that takes into account disease complexity.