flye ultra slow

[asogg]

Hello,
I am using flye to assemble a streptococcus agalatiae genome sequenced with minion mk1c and basecalled with dorado 0.8.2 in hac mode.
the program seems to crash on the cluster , even checking the processes it is using only 1 cpu, and after more than 2h it still has not finished lassembly of a single file . i attach the log file .
any suggestions about this?
thanks
Alessio

flye.log

[mikolmogorov]

It is definitely unusually slow. Did the run end up progressing? It looks like you are using error corrected mode, but if this is raw data you should use either --nano-hq or --nano-raw, depending on the chemistry (see the manual for details).

[asogg]

i changed with nano-hq but failed too. i used the --meta option and it worked but lot of short contigs so i moved to the asm option at 50x with nano-hq and it was fine . the problem maybe was the coverage , near 700X. feel free to close the issue.
thanks.
Alessio

[asogg]

Hello, attached there is the log file of the successful flye run . i think it is not junk (is a pure bacterial isolate) . we have to test the other two assemblers , were are waiting to install on the cluster and after we start testing. if you need more info let me know. regards Alessio Alessio Soggiu, PhD Associate professor, Lab. Microbial Genomics and Proteomics Department of Biomedical, Surgical and Dental Sciences (DSBCO) One-Health unit University of Milan. Via Pascal.36, 20133 Italy. Phone +39 02.503.15138 Secretary Italian Proteomics Association (ItPA) www.itpa.it<http://www.itpa.it/> Committee member European Bioinformatics Community (EuBIC) https://www.proteomics-academy.org/ Dental and Maxillo-Facial Surgery Unit at Policlinico Hospital Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico di Milan Via della Commenda 10 - 20122 Milano – Italia

…

________________________________ Da: John Urban ***@***.***> Inviato: venerdì, gennaio 24, 2025 3:59 PM A: mikolmogorov/Flye ***@***.***> Cc: Alessio Soggiu ***@***.***>; Author ***@***.***> Oggetto: Re: [mikolmogorov/Flye] flye ultra slow (Issue #744) Is there any possibility the dataset is junk? Have other assemblers given you trouble as well, or only Flye? Other lightweight ultra-fast assemblers you might consider are: * WtdBg2: https://github.com/ruanjue/wtdbg2 * Miniasm: https://github.com/lh3/miniasm Have you tried mapping the data to a previous "streptococcus agalatiae" assembly? It seems many are located here: https://www.ncbi.nlm.nih.gov/datasets/genome/?taxon=1311 I'm just curious as to the quality of your data. What percent of reads map to the other assemblies, and with what accuracy? If it is a low percent mapping, you might consider trying something like Kraken or simply BLASTing a sample of reads to NT to get an idea of what contamination sources are present, if any. — Reply to this email directly, view it on GitHub<#744 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADYR64U6SHMXZ4SU6WPXVFT2MJINNAVCNFSM6AAAAABVSOZNCSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMJSG4ZTMMBWGM>. You are receiving this because you authored the thread.