The aim is to make significant progress toward the reconstitution of mitotic spindle in vitro. We recently managed to reconstitute the self-organization of microtubules by both plus-end and minus-end directed motors. To our surprise, we found that microtubules and motors could self-pattern themselves into multiple domains, containing either plus-end or minus-end directed motors, separated by bundles of aligned microtubules (Utzschneider et al. PNAS, 2025). However, in these conditions, microtubules were stabilized in order to facilitate the transport of motors. We propose to further explore conditions closer to mitotic spindles by studying how motors of opposite polarities could self-organize dense arrays of short and dynamic microtubules. Students will learn the use of purified tubulin and molecular motors to grow dynamic networks of microtubules in vitro. They will also learn various surface treatments and microfluidic methods to confine and encapsulate these networks. They will be trained on first-class microscopes to monitor network self-organization. An ongoing with Jean-François Joanny (Collège de France) will allow the design of a physical model of active phase separation in the self-organization of mitotic spindles.