Localisation in a quasi-category  Let $W$ be a family of arrows in a category $\mathcal{C}$, there is a nature notion of localisation w.r.t. $W$. And if $W$ satisfies some nice properties, we have calculus of fraction. 
Now consider a quasi-category(i.e., weak Kan complex) $\mathcal{C}$, in which every $k$-arrows are invertible if $k>1$. Given a family of $1$-arrows $W$, is there a notion of localisation w.r.t. $W$, that is a universal way to invert all arrows in $W$? Furthermore, if $W$ is nice enough, does the notion of the calculus of fraction have a sensible generalization?
When I search localisation + quasicategory online, it yields mostly simplicial Localisation, Hammock localisation...which state how to localise a category rather than a quasi-catgory.
 A: You asked about a category of fractions, so it sounds  to me like you really want to formally invert the arrows in $W$. For that, you want something different than what Urs is explaining, however, there is still a way to do this hidden in HTT. Given a quasicategory $C$ with a class $W$ of morphisms, consider the pair $\left(C,W\right)$ as a marked simplicial set. The model category of marked simplicial sets is Quillen equivalent to simplicial sets with the Joyal model structure, and the fibrant objects are of the form $\left(C,inv.\right)$ where $C$ is a quasicategory $inv.$ stands for the class of all equivalences in $C$. The fibrant replacement of $\left(C,W\right)$ is $\left(C[W^{-1}],inv.\right)$- the "homotopy category" of $C$ you are searching for.
A: Some aspects and plenty of further pointers (mostly to Jacob Lurie's book, of course) are collected here:
http://ncatlab.org/nlab/show/localization+of+an+(infinity,1)-category
http://ncatlab.org/nlab/show/reflective+sub-(infinity,1)-category
