Max memory limit when running CITE-seq-Count

[molybdenumt]

I'm having issue running CITE-seq-Count because it's taking up too much memory. I tried running it on a cluster and even that got timed out. Start Time = 12/06/2021 23:26:26.956 End Time = 12/06/2021 23:34:26.087 CPU = 00:57:06 Max vmem = 189.568G Max rss = NA failed execd enforced h_vmem limit because: job 761625.1 died through signal KILL (9)

I have a set of 4 x R1 fastq files and 4 x R2 fastq files. Do you think it would use less memory if I try to aggregate the fastq files into 1 R1 fastq file and run them like that instead of listing all 4xfastq files? Anyone have ideas on how I can use less memory to run CITE-seq-Count?

[emjbishop]

I think if you give it fewer cores (--threads) it takes less memory. Could you share the command you used to run it? I've been able to run it with 32GB of RAM by only giving it one core (takes all night though). That was also with just 1 x R1 and 1 x R2 fastqs.