TypeError numpy.float

[brambloemen]

Hello,

I would very much like to use this tool for binning my Flye metagenome assemblies.
However, I ran into the following error:

graphmb --assembly Flye/ --outdir GraphMB --depth depths.txt --numcores 8

logging to GraphMB/20240624-090003graphmb_output.log
Running GraphMB 0.2.5
using cuda: False
setting seed to 1
setting tf seed
Reading cache from
Not using SCG file: marker_gene_stats.tsv (not found)
Reading assembly info file
found circular contig contig_140 mult. 1
--- name depth markers label edges
found circular contig contig_877 mult. 1
--- name depth markers label edges
found circular contig contig_903 mult. 1
--- name depth markers label edges
found circular contig contig_897 mult. 1
--- name depth markers label edges
==============================
DATASET STATS:
number of sequences: 594
assembly length: 0.031 Gb
assembly N50: 0.152 Mb
assembly average length (Mb): 0.053 max: 4.914 min: 0.0
Uncaught exception
Traceback (most recent call last):
  File "/data/brbloemen/mambaforge/bin/graphmb", line 8, in <module>
    sys.exit(main())
  File "/data/brbloemen/mambaforge/lib/python3.10/site-packages/graphmb/main.py", line 480, in main
    dataset.print_stats()
  File "/data/brbloemen/mambaforge/lib/python3.10/site-packages/graphmb/contigsdataset.py", line 177, in print_stats
    print("coverage samples: {}".format(len(self.node_depths[0])))
TypeError: object of type 'numpy.float64' has no len()
srun: error: bioit-rd: task 0: Exited with exit code 1

[AndreLamurias]

Hi, what dataset are you using and what's the content of the depths.txt file?

[brambloemen]

Hi,

I'm trying to run the tool on a metagenomic assembly of nanopore reads generated with Flye v2.9.3

I'm using the following command to generate the depths.txt:
jgi_summarize_bam_contig_depths --outputDepth {output.txt} {input.bam}
The input for this is a bam file generated by aligning nanopore reads (fastq) to a Flye --meta assembly using minimap2.

Here is the depth file:
depths.txt

[AndreLamurias]

Your file seems fine so it's quite possible that it's not being read by GraphMB
The file should be inside Flye/
Can you check GraphMB/20240624-090003graphmb_output.log for any message about the depths file?

[brambloemen]

I did copy the file to the Flye/ dir.
Here is the log:

Running GraphMB 0.2.5
using cuda: False
setting seed to 1
Reading cache from
Not using SCG file: marker_gene_stats.tsv (not found)
Reading assembly info file
Uncaught exception
Traceback (most recent call last):
  File "/data/brbloemen/mambaforge/bin/graphmb", line 8, in <module>
    sys.exit(main())
  File "/data/brbloemen/mambaforge/lib/python3.10/site-packages/graphmb/main.py", line 480, in main
    dataset.print_stats()
  File "/data/brbloemen/mambaforge/lib/python3.10/site-packages/graphmb/contigsdataset.py", line 177, in print_stats
    print("coverage samples: {}".format(len(self.node_depths[0])))
TypeError: object of type 'numpy.float64' has no len()

[AndreLamurias]

Can you run again with the options: ""--loglevel debug --reload" ?

[brambloemen]

When rerunning with the options you mention, I seem to get this error instead:

Running GraphMB 0.2.5
Namespace(assembly='Flye/', assembly_name='assembly.fasta', graph_file='assembly_graph.gfa', edge_threshold=None, depth='depths.txt', features='features.tsv', labels=None, embs=None, model_name='gcn', activation='relu', layers_vae=2, layers_gnn=3, hidden_gnn=128, hidden_vae=512, embsize_gnn=32, embsize_vae=64, batchsize=256, batchtype='auto', dropout_gnn=0.1, dropout_vae=0.2, lr_gnn=0.01, lr_vae=0.001, graph_alpha=1, kld_alpha=200, ae_alpha=1, scg_alpha=1, clusteringalgo='vamb', kclusters=None, aggtype='lstm', decoder_input='vae', vaepretrain=500, ae_only=False, negatives=10, quick=False, classify=False, fanout='10,25', epoch=500, print=10, evalepochs=20, evalskip=50, eval_split=0.0, kmer=4, rawfeatures=False, clusteringloss=False, targetmetric='hq', concatfeatures=False, no_loss_weights=True, no_sample_weights=True, early_stopping=0.1, nruns=1, mincontig=1000, minbin=200000, mincomp=1, randomize=False, labelgraph=False, binarize=False, noedges=False, read_embs=False, reload=True, markers='marker_gene_stats.tsv', post='writeembs_contig2bin', writebins=False, skip_preclustering=False, outname='graphmb', cuda=False, noise=False, savemodel=False, tsne=False, numcores=8, outdir='GraphMB', assembly_type='flye', contignodes=False, seed=1, quiet=False, read_cache=False, version=False, loglevel='debug')
using cuda: False
setting seed to 1
Cache not found on GraphMB
processing sequences Flye/assembly.fasta
read 594 seqs
processing GFA file (edge nodes) Flye/assembly_graph.gfa
skipped contigs 904 < 1000
read 0, edges
reading depths
reading labels
Saved cache to GraphMB

Not using SCG file: marker_gene_stats.tsv (not found)
Reading assembly info file
==============Running VAE model=====================
Uncaught exception
Traceback (most recent call last):
  File "/data/brbloemen/mambaforge/envs/graphmb/bin/graphmb", line 8, in <module>
    sys.exit(main())
  File "/data/brbloemen/mambaforge/envs/graphmb/lib/python3.9/site-packages/graphmb/main.py", line 499, in main
    vae_embs, _ = train_ccvae.run_model_ccvae(dataset, args, logger, 0,
  File "/data/brbloemen/mambaforge/envs/graphmb/lib/python3.9/site-packages/graphmb/train_ccvae.py", line 154, in run_model_ccvae
    X, adj, cluster_mask, neg_pair_idx, pos_pair_idx = prepare_data_for_gnn(
  File "/data/brbloemen/mambaforge/envs/graphmb/lib/python3.9/site-packages/graphmb/train_ccvae.py", line 91, in prepare_data_for_gnn
    edge_features = edge_weights / edge_weights.max()
  File "/data/brbloemen/mambaforge/envs/graphmb/lib/python3.9/site-packages/numpy/core/_methods.py", line 40, in _amax
    return umr_maximum(a, axis, None, out, keepdims, initial, where)
ValueError: zero-size array to reduction operation maximum which has no identity