UFold
UFold copied to clipboard
Having trouble installing on MacOS
Hello, I know this is probably not a bug with the code but I was hoping someone here would know.
I intend to run this on a M1 Mac and I cannot get past the conda env create -f UFold.yml A few things stand out for me. One is that it seems like conda cannot resolve the packages in that yml, but I also don't know much of Conda, so I might be mis-interpreting what this does.
Second is that I see CUDA references, but I don't have an NVDIA card. Will this still run on CPU or Apple GPU perhaps? If so, should I remove the CUDA references or that is a show stopper for me?
I've been working on a project to slice motifs from RNA and I was hoping I could tailor my output to fit UFold and script it so it goes out of our tool and into UFold and make it a breeze for users to input a sequence and get all the folded structures.
Just in case, this is what I'm getting from the command:
conda env create -f UFold.yml Collecting package metadata (repodata.json): done Solving environment: failed
ResolvePackageNotFound:
- cudatoolkit==10.1.243=h036e899_8
- readline==7.0=hb321a52_4
- ninja==1.10.2=hff7bd54_1
- libopenblas==0.3.13=h4367d64_0
- libffi==3.2.1=hf484d3e_1007
- _openmp_mutex==4.5=1_gnu
- libgomp==9.3.0=h5101ec6_17
- libgfortran4==7.5.0=ha8ba4b0_17
- sqlite==3.26.0=h7b6447c_0
- setuptools==52.0.0=py36h06a4308_0
- python==3.6.6=hd21baee_1003
- tk==8.6.10=hbc83047_0
- pip==21.2.2=py36h06a4308_0
- zlib==1.2.11=h7b6447c_3
- scipy==1.3.1=py36he2b7bc3_0
- ca-certificates==2021.7.5=h06a4308_1
- libgcc-ng==9.3.0=h5101ec6_17
- libstdcxx-ng==11.1.0=h56837e0_8
- pandas==0.25.0=py36he6710b0_0
- pytorch==1.4.0=py3.6_cuda10.1.243_cudnn7.6.3_0
- xz==5.2.5=h7b6447c_0
- libgfortran-ng==7.5.0=ha8ba4b0_17
- mkl==2021.3.0=h06a4308_520
- openssl==1.0.2u=h7b6447c_0
- certifi==2021.5.30=py36h06a4308_0
- intel-openmp==2021.3.0=h06a4308_3350
Hello nlabrad, I am a MacUser and I encounntered the same problem as yours. But I ran it successfully on Linux system, I think this is a software intended for Linux only. You may try other RNA secondary structure predictor(for instance, RNAfold in ViennaRNA,I installed it successfully on my mac)