Spaces:
Starting
on
A10G
Apply for community grant: Academic project
DiffLinker is a an E(3)-equivariant diffusion model for molecular linker design. Given a set of disconnected fragments in 3D, DiffLinker places missing atoms in between and designs a molecule incorporating all the initial fragments. Our method can link an arbitrary number of fragments, requires no information on the attachment atoms and linker size, and can be conditioned on the protein pockets.
Right now, DiffLinker is running on basic CPU and can be used only for demo runs. I would like to apply for a community GPU grant to provide practitioners with the fast and efficient tool for addressing various drug discovery problems, where fragment linking is a key challenge. With GPU (ideally, 32GB RAM) hardware, it will be possible to allow users to run DiffLinker conditioned on a target protein pocket which is currently impossible. Besides, it will be possible to get more samples in a shorter time.
Please find our paper here: https://arxiv.org/abs/2210.05274
Hi @igashov , we have assigned a gpu to this space. Note that GPU Grants are provided temporarily and might be removed after some time if the usage is very low.
To learn more about GPUs in Spaces, please check out https://huggingface.co/docs/hub/spaces-gpus, this is a10g, 24gb vram is it possible to apply some optimizations to get it working on a10g?