• Adds support for the new PAE JSON format while also keeping backwards compatibility with the legacy format.
• The new format makes the PAE JSON about 4x smaller when stored compressed, even smaller when stored uncompressed (we don't store redundant residue indices and the predicted aligned error is rounded to integers).
• The new format parses about 3x faster.
• I kept the dtype of the matrix as np.float64 to not break any existing code, but since PAE is now integers, np.int32, np.float32 or even np.float16 could be used is RAM usage is too high.

Hi,

Thanks for these scripts, seems very useful. However, I have some problem running it. I set up a colab notebook with copy-pasting and editing the code so it is runnable without cmd arguments. I included the same example as you used in the readme.

https://colab.research.google.com/drive/1xWsXMMolZqAL5bIAvl2pAqNuOV_Q0V0l?usp=sharing

If I use networkx, I get the following error:

TypeError                                 Traceback (most recent call last)

[<ipython-input-8-4af6c93a04f6>](https://localhost:8080/#) in <module>
----> 1 clusters = f(pae, pae_power=pae_power, pae_cutoff=pae_cutoff, graph_resolution=resolution)
      2 max_len = max([len(c) for c in clusters])
      3 clusters = [list(c) + ['']*(max_len-len(c)) for c in clusters]
      4 output_file = output_file

1 frames

[/usr/local/lib/python3.7/dist-packages/networkx/algorithms/community/modularity_max.py](https://localhost:8080/#) in greedy_modularity_communities(G, weight, resolution, n_communities)
    111     for u, nbrdict in dq_dict.items():
    112         for v, wt in nbrdict.items():
--> 113             dq_dict[u][v] = q0 * wt - resolution * (a[u] * b[v] + b[u] * a[v])
    114 
    115     # Use -dq to get a max_heap instead of a min_heap

TypeError: can't multiply sequence by non-int of type 'numpy.float64'

When I try igraph, I get this:

/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:80: RuntimeWarning: divide by zero encountered in true_divide

---------------------------------------------------------------------------

TypeError                                 Traceback (most recent call last)

[<ipython-input-11-4af6c93a04f6>](https://localhost:8080/#) in <module>
----> 1 clusters = f(pae, pae_power=pae_power, pae_cutoff=pae_cutoff, graph_resolution=resolution)
      2 max_len = max([len(c) for c in clusters])
      3 clusters = [list(c) + ['']*(max_len-len(c)) for c in clusters]
      4 output_file = output_file

[<ipython-input-2-c5063efac596>](https://localhost:8080/#) in domains_from_pae_matrix_igraph(pae_matrix, pae_power, pae_cutoff, graph_resolution)
     88     g.es['weight']=sel_weights
     89 
---> 90     vc = g.community_leiden(weights='weight', resolution_parameter=graph_resolution/100, n_iterations=-1)
     91     membership = numpy.array(vc.membership)
     92     from collections import defaultdict

TypeError: unsupported operand type(s) for /: 'tuple' and 'int'

If I try igraph and set the diagonal to 1, I still get the same error, only this line is missing from the above error: /usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:80: RuntimeWarning: divide by zero encountered in true_divide

Do you have any idea what might cause the problem? Thank you for your help!

Hi,

I was playing around with the different threshold parameters and realised that the same run can yield different number of clusters.

python pae_to_domains.py *_predicted_aligned_error_v1.json --pae_cutoff 10 --pae_power 1 --resolution 1
Wrote 13 clusters to clusters.csv. Biggest cluster contains 54 residues. Run time was 0.86 seconds.

python pae_to_domains.py *_predicted_aligned_error_v1.json --pae_cutoff 10 --pae_power 1 --resolution 1
Wrote 14 clusters to clusters.csv. Biggest cluster contains 54 residues. Run time was 0.73 seconds.

Is this something you also observed on the same input pae json file? I suspect this comes from the igraph community_leiden step, but I might be wrong.