Edge weights must not be NaN values. -- Invalid value

[fgonzalez3]

Hi there! I am running the below Snakemake workflow to bin some short read assemblies. I'm testing my workflow on two samples before I run it on my entire sample set, but I am encountering the error described in the title for one of those samples. Below, I've written out the workflow, in addition to the error. I haven't seen this issue brought up so far, so I'm hoping to get some feedback on where I could be producing the corrupted file in question. I'm happy to provide the preceding steps in the workflow if it would be of any aid. Thank you!

Workflow -

rule semibin2_generate_concatenated_db_spades:
    """
    Generate concatenated FASTA file necessary for SembiBin's multi-sample binning pipeline
    """
    input:
        contigs = expand("results/{genera}/3_dedup_contigs/SPAdes/individual_metagenome_assembly/{sample}/{sample}_DEDUP95.fasta", genera=config["genera"], sample=SAMPLES)
    output:
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa"
    params:
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db",
        threads = 4
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenate_fa.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenate_fa.err"
    shell:
        """
        module unload miniconda
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        SemiBin2 concatenate_fasta \
        --input-fasta {input.contigs} \
        --output {params.outdir} --compression=none
        """

rule sembin2_align_to_concatenated_db_spades:
    """
    Align reads from each sample to our concatenated FASTA db, necessary for SemiBin pipeline
    """
    input:
        db = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa",
        r1 = "results/{genera}/1_pre_processing/dedup_reads/{sample}/{sample}_host_removed_dedup_R1.fastq",
        r2 = "results/{genera}/1_pre_processing/dedup_reads/{sample}/{sample}_host_removed_dedup_R2.fastq"
    output:
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.1.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.2.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.3.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.4.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.rev.1.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.rev.2.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aligned_sorted.bam"
    params:
        genera=config["genera"],
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}"
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aln.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aln.err"
    shell:
        """
        module unload miniconda 
        module load Bowtie2/2.5.1-GCC-12.2.0
        module load SAMtools/1.21-GCC-12.2.0

        # 1. Build db index
        bowtie2-build \
        -f {input.db} {params.outdir}/{wildcards.sample}_indexed_contig \
        1>> {log.stdout} 2>> {log.stderr}

        # 2. Align reads back to SemiBin db index
        bowtie2 \
        -x {params.outdir}/{wildcards.sample}_indexed_contig -1 {input.r1} -2 {input.r2} | samtools view -b -F 4 -F 2048 | samtools sort -o {output[6]} \
        1>> {log.stdout} 2>> {log.stderr}
        """

rule semibin2_features_and_model_spades:
    """
    Generate sequence features and train model for SemiBin2
    """
    input:
        cat_fa = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa",
        bams = expand("results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aligned_sorted.bam", genera=config["genera"], sample=SAMPLES)
    output:
        split = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/data_split.csv",
        csv = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/data.csv"
    params:
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model",
        outdir2 =  "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}",
        threads = 4
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/sequence_features.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/sequence_features.err"
    shell:
        """
        module unload miniconda
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        # Generate sequence features data.csv & data_split.csv files
        SemiBin2 generate_sequence_features_multi \
        -i {input.cat_fa} \
        -b {input.bams} \
        -o {params.outdir} \
        -t {params.threads} \
        1>> {log.stdout} 2>> {log.stderr}

        mv {params.outdir}/samples/{wildcards.sample}_DEDUP95/data.csv {params.outdir2}
        mv {params.outdir}/samples/{wildcards.sample}_DEDUP95/data_split.csv {params.outdir2}
        """

rule semibin2_train_model_spades:
    """
    Train ML model on previously curated SemiBin2 feature data
    """
    input:
        split = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/data_split.csv",
        csv = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/data.csv"
    output:
        model = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/train_model/{sample}/model.pt"
    params:
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/train_model/{sample}",
        threads = 4
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/train_model/{sample}/ML_train.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/train_model/{sample}/ML_train.err"
    shell:
        """
        module unload miniconda
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        # Train model
        SemiBin2 train_self \
        --data {input.csv} \
        --data-split {input.split} \
        -o {params.outdir} \
        -t {params.threads} \
        1>> {log.stdout} 2>> {log.stderr}
        """

checkpoint semibin2_bin_spades:
    """
    Bin contigs using SemiBin's multi-sample binning model for individual binning of samples.
    This method often returns the most complete bins and is most optimized for complex samples.    
    """
    input:
        contigs = "results/{genera}/3_dedup_contigs/SPAdes/individual_metagenome_assembly/{sample}/{sample}_DEDUP95.fasta",
        csv = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/features_and_model/{sample}/data.csv",
        model = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/train_model/{sample}/model.pt"
    output:
        outdir = directory("results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}/output_bins"),
        check = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}/contig_bins.tsv"
    params:
        base = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}",
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}/output_bins",
        minlen = "1500",
        threads = 4
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}/bin.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/{sample}/bin.err"
    shell:
        """
        module unload miniconda
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        # Bin
        SemiBin2 bin_short \
        -i {input.contigs} \
        --model {input.model} \
        --data {input.csv} \
        -o {params.base} \
        -t {params.threads} \
        --min-len={params.minlen} \
        1>> {log.stdout} 2>> {log.stderr}
        """

def get_semibin2_bin_spades(wc):
    ckpt = checkpoints.semibin2_bin_spades.get(genera=wc.genera, sample=wc.sample)
    import os, glob
    bins_dir = ckpt.output.outdir
    return sorted(glob.glob(os.path.join(bins_dir, "*.fa.gz")))

Error -

2025-10-09 13:20:22 login2.mccleary.ycrc.yale.edu SemiBin2[1617592] INFO Running SemiBin2 version 2.2.0
2025-10-09 13:20:24 login2.mccleary.ycrc.yale.edu SemiBin2[1617592] WARNING Did not detect GPU or CUDA was not installed/supported, using CPU.
2025-10-09 13:20:24 login2.mccleary.ycrc.yale.edu SemiBin2[1617592] INFO Start binning.
multiprocessing.pool.RemoteTraceback: 
"""
Traceback (most recent call last):
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/multiprocessing/pool.py", line 125, in worker
    result = (True, func(*args, **kwds))
                    ~~~~^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 14, in run_infomap1
    return g.community_infomap(edge_weights=edge_weights, vertex_weights=vertex_weights, trials=trials)
           ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/igraph/community.py", line 56, in _community_infomap
    membership, codelength = GraphBase.community_infomap(
                             ~~~~~~~~~~~~~~~~~~~~~~~~~~~^
        graph, edge_weights, vertex_weights, trials
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
igraph._igraph.InternalError: Error at src/community/infomap/infomap.cc:263: Edge weights must not be NaN values. -- Invalid value
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/bin/SemiBin2", line 10, in <module>
    sys.exit(main2())
             ~~~~~^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/main.py", line 1605, in main2
    binning_short(logger, args.data, args.minfasta_kb * 1000, binned_length,
    ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            environment=args.environment, contig_dict=contig_dict,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            model_path=args.model_path, output=args.output,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            device=device, args=args)
            ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/main.py", line 1222, in binning_short
    cluster(
    ~~~~~~~^
        logger,
        ^^^^^^^
    ...<8 lines>...
        binned_length=binned_length,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        minfasta=minfasta)
        ^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 289, in cluster
    embedding, contig_labels = run_embed_infomap(logger, model, data,
                               ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^
            device=device, max_edges=args.max_edges, max_node=args.max_node,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            is_combined=is_combined, n_sample=n_sample,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            contig_dict=contig_dict, num_process=args.num_process,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            random_seed=args.random_seed)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 182, in run_embed_infomap
    result = run_infomap(g,
                edge_weights=edge_weights,
                vertex_weights=length_weight,
                num_process=num_process)
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 26, in run_infomap
    rs = [r.get() for r in rs]
          ~~~~~^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/multiprocessing/pool.py", line 774, in get
    raise self._value
igraph._igraph.InternalError: Error at src/community/infomap/infomap.cc:263: Edge weights must not be NaN values. -- Invalid value

[fgonzalez3]

I receive the same error when running it on easy mode:

**Workflow - **

rule semibin2_generate_concatenated_db_spades:
    """
    Generate concatenated FASTA file necessary for SembiBin's multi-sample binning pipeline
    """
    input:
        contigs = expand("results/{genera}/3_dedup_contigs/SPAdes/individual_metagenome_assembly/{sample}/{sample}_DEDUP95.fasta", genera=config["genera"], sample=SAMPLES)
    output:
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa"
    params:
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db",
        threads = 4
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenate_fa.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenate_fa.err"
    shell:
        """
        module unload miniconda
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        SemiBin2 concatenate_fasta \
        --input-fasta {input.contigs} \
        --output {params.outdir} --compression=none
        """

rule sembin2_align_to_concatenated_db_spades:
    """
    Align reads from each sample to our concatenated FASTA db, necessary for SemiBin pipeline
    """
    input:
        db = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa",
        r1 = "results/{genera}/1_pre_processing/dedup_reads/{sample}/{sample}_host_removed_dedup_R1.fastq",
        r2 = "results/{genera}/1_pre_processing/dedup_reads/{sample}/{sample}_host_removed_dedup_R2.fastq"
    output:
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.1.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.2.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.3.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.4.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.rev.1.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}/{sample}_indexed_contig.rev.2.bt2",
        "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aligned_sorted.bam"
    params:
        genera=config["genera"],
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}"
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aln.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aln.err"
    shell:
        """
        module unload miniconda 
        module load Bowtie2/2.5.1-GCC-12.2.0
        module load SAMtools/1.21-GCC-12.2.0

        # 1. Build db index
        bowtie2-build \
        -f {input.db} {params.outdir}/{wildcards.sample}_indexed_contig \
        1>> {log.stdout} 2>> {log.stderr}

        # 2. Align reads back to SemiBin db index
        bowtie2 \
        -x {params.outdir}/{wildcards.sample}_indexed_contig -1 {input.r1} -2 {input.r2} | samtools view -b -F 4 -F 2048 | samtools sort -o {output[6]} \
        1>> {log.stdout} 2>> {log.stderr}
        """

rule easy_multi_bin:
    """
    Run Semibin2 on easy multi binning mode
    """
    input:
        fa = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/generate_concatenated_db/concatenated.fa",
        bams = expand("results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/{sample}_aligned_sorted.bam", genera=config["genera"], sample=SAMPLES)
    output:
        outdir = directory("results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/output_bins"),
        check = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/check.tsv"
    params:
        base = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/",
        outdir = "results/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/output_bins",
        compress = "none",
        minlen = 1500,
        threads = 1
    log:
        stdout = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/multi_binning.out",
        stderr = "logs/{genera}/6_binning/semibin2/SPAdes_individual_assembly/binning/multi_binning.err"
    shell:
        """
        module unload miniconda 
        source activate /vast/palmer/pi/turner/flg9/conda_envs/semibin

        SemiBin2 multi_easy_bin \
        -i {input.fa} \
        -b {input.bams} \
        --self-supervised \
        --min-len {params.minlen} \
        --threads {params.threads} \
        -o {params.base} \
        --compression {params.compress} \
        1>> {log.stdout} 2>> {log.stderr}

        ls {params.outdir}/*.fa > {output.check}
        """

def get_semibin2_bin_spades(wc):
    ckpt = checkpoints.easy_multi_bin.get(genera=wc.genera, sample=wc.sample)
    import os, glob
    bins_dir = ckpt.output.outdir
    return sorted(glob.glob(os.path.join(bins_dir, "*.fa")))

**Error - **

2025-10-14 14:33:16 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Running SemiBin2 version 2.2.0
2025-10-14 14:33:16 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Binning for short_read
2025-10-14 14:33:21 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] WARNING Did not detect GPU or CUDA was not installed/supported, using CPU.
2025-10-14 14:33:21 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Performing multi-sample binning
2025-10-14 14:33:21 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Generating training data...
2025-10-14 14:33:23 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Calculating coverage for every sample.
2025-10-14 14:34:36 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Processed: results/beluga/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/Sample_JUNO_03AUG2023_aligned_sorted.bam
2025-10-14 14:35:24 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Processed: results/beluga/6_binning/semibin2/SPAdes_individual_assembly/align_to_concatenated_db/Sample_JUNO_04JUN2024_aligned_sorted.bam
2025-10-14 14:35:33 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Training model and clustering for sample "Sample_JUNO_03AUG2023_DEDUP95"
2025-10-14 14:35:33 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Start training from a single sample.
2025-10-14 14:35:33 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Training model...

  0%|          | 0/15 [00:00<?, ?it/s]
  7%|▋         | 1/15 [00:42<09:48, 42.06s/it]
 13%|█▎        | 2/15 [01:24<09:07, 42.14s/it]
 20%|██        | 3/15 [02:06<08:25, 42.13s/it]
 27%|██▋       | 4/15 [02:48<07:42, 42.05s/it]
 33%|███▎      | 5/15 [03:29<06:59, 41.91s/it]
 40%|████      | 6/15 [04:12<06:19, 42.16s/it]
 47%|████▋     | 7/15 [04:55<05:38, 42.26s/it]
 53%|█████▎    | 8/15 [05:37<04:56, 42.32s/it]
 60%|██████    | 9/15 [06:19<04:12, 42.15s/it]
 67%|██████▋   | 10/15 [07:01<03:30, 42.05s/it]
 73%|███████▎  | 11/15 [07:42<02:47, 41.98s/it]
 80%|████████  | 12/15 [08:24<02:05, 41.82s/it]
 87%|████████▋ | 13/15 [09:05<01:23, 41.71s/it]
 93%|█████████▎| 14/15 [09:48<00:41, 41.92s/it]
100%|██████████| 15/15 [10:29<00:00, 41.75s/it]
100%|██████████| 15/15 [10:29<00:00, 41.97s/it]
2025-10-14 14:46:08 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Training finished.
2025-10-14 14:46:08 r813u03n09.mccleary.ycrc.yale.edu SemiBin2[3541388] INFO Start binning.
/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py:157: RuntimeWarning: invalid value encountered in multiply
  kl *= -1
/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py:160: RuntimeWarning: invalid value encountered in subtract
  kl_matrix -= kl
Traceback (most recent call last):
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/bin/SemiBin2", line 10, in <module>
    sys.exit(main2())
             ~~~~~^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/main.py", line 1635, in main2
    multi_easy_binning(
    ~~~~~~~~~~~~~~~~~~^
        logger,
        ^^^^^^^
        args,
        ^^^^^
        device)
        ^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/main.py", line 1382, in multi_easy_binning
    binning_short(**binning_kwargs)
    ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/main.py", line 1222, in binning_short
    cluster(
    ~~~~~~~^
        logger,
        ^^^^^^^
    ...<8 lines>...
        binned_length=binned_length,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        minfasta=minfasta)
        ^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 289, in cluster
    embedding, contig_labels = run_embed_infomap(logger, model, data,
                               ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^
            device=device, max_edges=args.max_edges, max_node=args.max_node,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            is_combined=is_combined, n_sample=n_sample,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            contig_dict=contig_dict, num_process=args.num_process,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            random_seed=args.random_seed)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 182, in run_embed_infomap
    result = run_infomap(g,
                edge_weights=edge_weights,
                vertex_weights=length_weight,
                num_process=num_process)
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/SemiBin/cluster.py", line 19, in run_infomap
    return g.community_infomap(edge_weights=edge_weights, vertex_weights=vertex_weights, trials=NR_INFOMAP_TRIALS)
           ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/vast/palmer/pi/turner/flg9/conda_envs/semibin/lib/python3.13/site-packages/igraph/community.py", line 56, in _community_infomap
    membership, codelength = GraphBase.community_infomap(
                             ~~~~~~~~~~~~~~~~~~~~~~~~~~~^
        graph, edge_weights, vertex_weights, trials
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
igraph._igraph.InternalError: Error at src/community/infomap/infomap.cc:263: Edge weights must not be NaN values. -- Invalid value

[paulzierep]

We also have this error frequently:

2025-11-18 13:15:32 c192m1536-n3712 SemiBin[628433] INFO Binning for short_read
2025-11-18 13:15:52 c192m1536-n3712 SemiBin[628433] INFO Did not detect GPU, using CPU.
2025-11-18 13:15:54 c192m1536-n3712 SemiBin[628433] INFO Generating training data...
2025-11-18 13:16:01 c192m1536-n3712 SemiBin[628433] INFO Calculating coverage for every sample.
2025-11-18 13:17:54 c192m1536-n3712 SemiBin[628433] INFO Processed: ERR2231568.bam
2025-11-18 13:17:56 c192m1536-n3712 SemiBin[628433] INFO Start binning.
multiprocessing.pool.RemoteTraceback: 
"""
Traceback (most recent call last):
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/multiprocessing/pool.py", line 125, in worker
    result = (True, func(*args, **kwds))
                    ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/cluster.py", line 14, in run_infomap1
    return g.community_infomap(edge_weights=edge_weights, vertex_weights=vertex_weights, trials=trials)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/igraph/community.py", line 56, in _community_infomap
    membership, codelength = GraphBase.community_infomap(
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
igraph._igraph.InternalError: Error at src/community/infomap/infomap.cc:263: Edge weights must not be NaN values. -- Invalid value
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/bin/SemiBin2", line 10, in <module>
    sys.exit(main2())
             ^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/main.py", line 1600, in main2
    single_easy_binning(
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/main.py", line 1301, in single_easy_binning
    binning_short(**binning_kwargs)
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/main.py", line 1208, in binning_short
    cluster(
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/cluster.py", line 281, in cluster
    embedding, contig_labels = run_embed_infomap(logger, model, data,
                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/cluster.py", line 183, in run_embed_infomap
    result = run_infomap(g,
             ^^^^^^^^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/site-packages/SemiBin/cluster.py", line 26, in run_infomap
    rs = [r.get() for r in rs]
          ^^^^^^^
  File "/usr/local/tools/_conda/envs/__semibin@2.1.0/lib/python3.12/multiprocessing/pool.py", line 774, in get
    raise self._value
igraph._igraph.InternalError: Error at src/community/infomap/infomap.cc:263: Edge weights must not be NaN values. -- Invalid value

[hildebra]

Hey all,
Also running into this error with semibin2.2.1 (which is the latest I guess). Any chance that this can be fixed @luispedro ? Is this a problem with the python libraries?
best,
Falk

[hildebra]

Another variant of probably the same error:
python3.11/site-packages/SemiBin/cluster.py:138: RuntimeWarning: invalid value encountered in subtract
kl_matrix -= kl
double free or corruption (!prev)

[alarasutcu]

Hi all, I am also getting the same error when running the following command:

SemiBin2 single_easy_bin --self-supervised -i "{assembly}" -b "{bam_file}" -o "{semibin_output}" --compression none -t {threads} -m 2500 2> "{semibin_output}/semibin2.errlog"

I’ve been getting this error alot and I was able to bypass it by increasing the minimum size from 1500 to 2500, but I am still getting the error.