AttributeError: 'MaskedArray' object has no attribute 'folded'

[CoCo477]

Hi Dr. Portik,

I am running dadi for hypothesis testing with 5 2D models ("no_mig", "sec_contact_asym_mig", "anc_asym_mig", "sym_mig", "asym_mig") on 2 populations with
high-throughput sequencing data. I made the folded sfs files with easySFS from vcf files which I am then feeding to dadi_pipeline.

First, I used the scripts in Two_Population_Pipeline fold and made some necessary modifications to generate the Results_Summary_Short.txt file for each model. Then I'd like to use scripts in Goodness_of_Fit fold and make some necessary modifications to generate the Simulation_Results_[model name].txt file for each model. But there seems to be an issue with this step. Five models work perfectly fine, but one model(anc_asym_mig) keep failing with the following error:
Traceback (most recent call last):
File "Simulate_and_Optimize.py", line 232, in
param_labels=p_labels)
File "/data2/yangk/WYH/04.Demographic_History/03.LDfilter_dadi/ancient_gene_flow/Optimize_Functions_GOF.py", line 447, in Perform_S>
param_labels=param_labels, optimizer=optimizer)
File "/data2/yangk/WYH/04.Demographic_History/03.LDfilter_dadi/ancient_gene_flow/Optimize_Functions_GOF.py", line 253, in Optimize_>
output_file = "{}.log.txt".format(model_name))
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 742, in optimize_log_fmin
disp=False, maxiter=maxiter, full_output=True)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/scipy/optimize/optimize.py", line 543, in fmin
res = _minimize_neldermead(func, x0, args, callback=callback, **opts)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/scipy/optimize/optimize.py", line 689, in _minimize_neldermead
fsim[k] = func(sim[k])
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/scipy/optimize/optimize.py", line 427, in function_wrapper
return function(*(wrapper_args + args))
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 78, in _object_func_log
return _object_func(numpy.exp(log_params), args, **kwargs)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 54, in _object_func
result = ll_multinom(sfs, data)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 563, in ll_multinom
ll_arr = ll_multinom_per_bin(model, data)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 547, in ll_multinom_per_bin
return ll_per_bin(theta_optmodel, data)
File "/data2/yangk/miniconda3/envs/py36/lib/python3.6/site-packages/dadi/Inference.py", line 484, in ll_per_bin
if hasattr(data, 'folded') and data.folded and not model.folded:
AttributeError: 'MaskedArray' object has no attribute 'folded'

I found in dadi user group that someone also came across the same error using dadi_pipeline, but dadi's auther was not sure what exactly is happening (https://groups.google.com/g/dadi-user/c/-TXzZeo-TN0/m/70R0VvyhAAAJ).

I am attaching the errors files for anc_asym_mig model(error_for_anc_asym_mig.txt), the sfs files(fs_P2CSHJ64_P3CSHJ34.fs and max_fs_P2CSHJ86_P3CSHJ38.fs), the python script for running optimizations(dadi_Run_Optimizations_for_anc_asym_mig.py), the short summary results for 5 models' optimizations(Results_Summary_Short.txt), and the python script for simulation and optimization(Simulate_and_Optimize_for_anc_asym_mig.py). I would really appreciate some tips on how to make this model run in dadi_pipeline!

dadi_pipeline.zip

Thank you so much and all the best,

CoCo477