colab Umol "predict the protein-ligand complex structure" cell

[FJ0M]

I uploaded the .a3m file from HHBlits as outlined in the first cell, but run into this error?

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
[<ipython-input-8-55efa3eb56f8>](https://localhost:8080/#) in <cell line: 10>()
      8 
      9 #Predict
---> 10 predict(config.CONFIG,
     11             MSA_FEATS,
     12             LIGAND_FEATS,

7 frames
[/usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py](https://localhost:8080/#) in _create_c_op(graph, node_def, inputs, control_inputs, op_def, extract_traceback)
   1965   except errors.InvalidArgumentError as e:
   1966     # Convert to ValueError for backwards compatibility.
-> 1967     raise ValueError(e.message)
   1968 
   1969   # Record the current Python stack trace as the creating stacktrace of this

ValueError: Cannot reshape a tensor with 345790 elements to shape [2290,344,1] (787760 elements) for '{{node reshape_msa}} = Reshape[T=DT_INT32, Tshape=DT_INT32](Const_6, reshape_msa/shape)' with input shapes: [2290,151], [3] and with input tensors computed as partial shapes: input[1] = [2290,344,1].

The example in the notebook works fine, so it maybe my formatting, just thought I'd raise so you're aware. I'll try a local install and see if I can get past this.

[patrickbryant1]

Hi,
I think this is related to some shape mismatch between the MSA and other features. Please check that all lengths match.