Emu algorithm not finishing with short reads

Hi Kristen,

I've been using Emu to process some of my IonTorrent 16S reads. It works most of the time, but some samples do not finish processing. The output directory only contains a SAM file and no abundance table. This is the error message I get:

./HM_ion_16S_emu_emuDB_unfinished/AS16S_025_BC143 Unassigned read count: 12526 Assigned read count: 179918 [keshin@gra-login2 slurmout]$ cat ion-emu-emuDB_12381321_4294967294.err [M::mm_idx_gen::2.1231.25] collected minimizers [M::mm_idx_gen::2.2972.09] sorted minimizers [M::main::2.3102.08] loaded/built the index for 49243 target sequence(s) [M::mm_mapopt_update::2.3102.08] mid_occ = 1000 [M::mm_idx_stat] kmer size: 21; skip: 11; is_hpc: 0; #seq: 49243 [M::mm_idx_stat::2.3222.08] distinct minimizers: 625821 (48.62% are singletons); average occurrences: 19.472; average spacing: 6.083; total length: 74133434
[M::worker_pipeline::8.36016.92] mapped 192444 sequences [M::main] Version: 2.24-r1122 [M::main] CMD: minimap2 -ax sr -t 32 -N 50 -p .9 -K 500000000 -o ./HM_ion_16S_emu_emuDB_unfinished/AS16S_025_BC143/AS16S_025_BC143_filtered.fastq_emu_alignments.sam emu_database/species_taxid.fasta /home/keshin/scratch/unfinished_iontorrent/ion_emu_emudb/AS16S_025_BC143_filtered.fastq.gz [M::main] Real time: 8.437 sec; CPU: 141.525 sec; Peak RSS: 1.097 GB Traceback (most recent call last): File "./emu", line 805, in f_full, f_set_thresh, read_dist = expectation_maximization_iterations(log_prob_rgs, File "./emu", line 333, in expectation_maximization_iterations raise ValueError("total_log_likelihood decreased from prior iteration") ValueError: total_log_likelihood decreased from prior iteration

Is there any way to generate the abundance profiles from the SAM files?

Thank you, Kenny