google-deepmind / google-deepmind/alphafold
Different alphafold version get different results
- Dominant language
- Python
- Stars
- 14.9k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
I run the alphafold in different version(one is v2.2.4, another is earlier )and get different results using the model_preset of multimer. As shown in the picture:

This is a complex composed of Cas9 and SNRPD2 (red) proteins. The predicted struture of cas9 is nearly the same in both results, as well as SNRPD2. However, the binding site of the cas9 and SNRPD2 is different in two results.
I want to know if it's because of the different versions of alphafold or because of my command?
here is my command:
nohup python3 docker/run_docker.py \
--fasta_paths=/home/zwf/alphafold/hpl1.fasta,/home/zwf/alphafold/SNRPD2.fasta \
--model_preset=multimer \
--docker_user=0 \
--db_preset=full_dbs \
--output_dir=/data/alphafold_outdir \
--max_template_date=2021-12-10 \
--data_dir=/data/alphalfold &> /home/zwf/alphafold/SNRPD2.out&
Contributor guide
Assessment
This issue has not been assessed yet.