The LRRK2–ponatinib dataset was collected similarly, except for the following differences. Movies were recorded at defocus values from –0.6 to –1.8 μm at a magnification of 130 kx in hardware binning mode, corresponding to a pixel size of 0.6485 Å at the specimen. During 2.0 s exposure, 60 frames were collected with a total electron dose of ~68 e–/Å–2 (at a dose rate of 1.1 e–/frame/Å2). In total, 27,611 images were collected. Motion correction was performed on hardware-binned movie stacks and binned by 1 using MotionCor242 (link). CTF estimation was performed using Gctf43 (link). After the selection of high-quality micrographs, 24,254 images were used during the data process.
The LRRK2–ponatinib dataset was processed similarly in cryoSPARC. Briefly, particles were selected using the template picker and extracted using a binning factor of 4. Several rounds of the 2D classification were performed and two groups of good classes were observed, corresponding to the LRRK2 monomer and dimer states, respectively. Both groups were selected, and ab initio reconstruction was performed. Heterogeneous refinement was performed to further separate the LRRK2 monomer and dimer states. As a result, 95,805 particles were assigned to the monomer class and 75,849 particles to the dimer. Both 3D classes were further refined after extraction of unbinned particles corresponding to each identified sub-set. For the LRRK2 monomer state, we performed a standard NU-refinement without imposing symmetry. For the LRRK2 dimer state, we performed NU-refinement by applying C2 symmetry, and then symmetry expansion followed by focused refinement to further improve the resolution of each LRRK2 protomer without N-terminal ARM domain. All resolution estimates were calculated according to the gold-standard FSC using the 0.143 criterion45 (link). Local resolution was estimated in cryoSPARC. The density maps were B-factor sharpened in cryoSPARC and used to produce figures and build models.
The LRRK2–ponatinib dataset was processed similarly in cryoSPARC. Briefly, particles were selected using the template picker and extracted using a binning factor of 4. Several rounds of the 2D classification were performed and two groups of good classes were observed, corresponding to the LRRK2 monomer and dimer states, respectively. Both groups were selected, and ab initio reconstruction was performed. Heterogeneous refinement was performed to further separate the LRRK2 monomer and dimer states. As a result, 95,805 particles were assigned to the monomer class and 75,849 particles to the dimer. Both 3D classes were further refined after extraction of unbinned particles corresponding to each identified sub-set. For the LRRK2 monomer state, we performed a standard NU-refinement without imposing symmetry. For the LRRK2 dimer state, we performed NU-refinement by applying C2 symmetry, and then symmetry expansion followed by focused refinement to further improve the resolution of each LRRK2 protomer without N-terminal ARM domain. All resolution estimates were calculated according to the gold-standard FSC using the 0.143 criterion45 (link). Local resolution was estimated in cryoSPARC. The density maps were B-factor sharpened in cryoSPARC and used to produce figures and build models.
Free full text: Click here
