# Ab initio docking & unambiguous restraints **URL:** \ **Category:** HADDOCK\ **Created:** [June 16, 2021, 9:47am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113 "2021-06-16T09:47:51Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![jbibbe](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/jbibbe/32/432_2.png) [@jbibbe](https://ask.bioexcel.eu/u/jbibbe)\ **Post date:** [June 16, 2021, 9:47am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113/1 "2021-06-16T09:47:51Z")
I am doing the [tutorial about small molecule binding site screening](https://www.bonvinlab.org/education/HADDOCK24/HADDOCK24-binding-sites/) after having completed the [tutorial about using HADDOCK locally](https://www.bonvinlab.org/education/HADDOCK24/HADDOCK24-local-tutorial/). I want to do the small molecule binding site screening locally too, instead of through the server like in the tutorial. The target protein has three chains. In the tutorial for local HADDOCK, unambiguous chains were defined to keep the three chains together while docking. But the [HADDOCK manual says I cannot define unambiguous restraints while doing ab initio docking](https://www.bonvinlab.org/software/haddock2.4/airs/#random-air-definition-ab-initio-mode). Is there something else I can do to keep the three chains together in ab initio docking? ---
**Author:** ![amjjbonvin](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/amjjbonvin/32/23_2.png) [@amjjbonvin](https://ask.bioexcel.eu/u/amjjbonvin)\ **Post date:** [June 16, 2021, 10:07am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113/2 "2021-06-16T10:07:25Z")
> The target protein has three chains. In the tutorial for local HADDOCK, unambiguous chains were defined to keep the three chains together while docking. But the [HADDOCK manual says I cannot define unambiguous restraints while doing ab initio docking](https://www.bonvinlab.org/software/haddock2.4/airs/#random-air-definition-ab-initio-mode). Is there something else I can do to keep the three chains together in ab initio docking? In that tutorial only the rigid body docking models are analysed to identify putative binding sites and at that stage the three chains will not change. If you were to put those restraints in the hbonds category it should work though. ---
**Author:** ![jbibbe](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/jbibbe/32/432_2.png) [@jbibbe](https://ask.bioexcel.eu/u/jbibbe)\ **Post date:** [June 16, 2021, 10:26am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113/3 "2021-06-16T10:26:18Z")
Thanks, I now understand why it doesn’t matter that there are 3 chains and no unambiguous restraints. > If you were to put those restraints in the hbonds category it should work though. But the manual says: > **Note:** Random AIRs are exclusive with ambiguous, unambiguous and hydrogen bond restraints defined in new.html. They can however be combined with surface and center of mass restraints (see below). Doesn’t that mean that I cannot use hbonds, but I can use surface and center of mass restraints? ---
**Author:** ![amjjbonvin](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/amjjbonvin/32/23_2.png) [@amjjbonvin](https://ask.bioexcel.eu/u/amjjbonvin)\ **Post date:** [June 16, 2021, 10:37am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113/4 "2021-06-16T10:37:40Z")
hbonds should work - we need to correct the manual then (but do check the new 2.4 version of the manual). And rather use the new 2.4 version of haddock ---
**Author:** ![amjjbonvin](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/amjjbonvin/32/23_2.png) [@amjjbonvin](https://ask.bioexcel.eu/u/amjjbonvin)\ **Post date:** [June 16, 2021, 10:46am UTC](https://ask.bioexcel.eu/t/ab-initio-docking-unambiguous-restraints/3113/5 "2021-06-16T10:46:07Z")
Actually random AIRs are only applied in the rigid body docking part (it0) and replaced by surface restraints for the flexible refinement. As such ambiguous and unambiguous restraints can be defined and will be used in it1 But they should only contain intramolecular restraints otherwise it does not make sense.