Running metaSPAdes with 71M paired-end reads, advice on RAM and command parameters

[suma-h-bio]

Hello,

I am a beginner in shotgun metagenomic analysis, and I am currently running metaSPAdes on my dataset.

Input size: ~71,251,478 paired-end reads (1 sample)

System specs: 125 GB RAM available + 2 TB swap memory

Command used:
metaspades.py -1 sample_R1.fastq.gz -2 sample_R2.fastq.gz
-k 21,33,55,77,99,127 -t 16 -m 250

I have a couple of doubts:

1. Is it okay that I did not specify --only-assembler (I ran with default error correction)?

2. Will metaSPAdes run correctly with my specs, or should I adjust -m, k-mer values, or other parameters?

3. Since I am new to shotgun metagenomics, do you recommend any optimized settings for large datasets like mine?

Any guidance will be very helpful 🙏

Thank you!