ballooning assembly as I add more coverage

[waalkes]

I am paired end assembling a known genome - Salmonella enterica strain LT2 (CP014051.2) ~4.9M in size

I am assembling at 20x,50x,100x,300x and 600x coverage.

The results I am getting are strange. I am assembling with default parameters:
./spades.py --pe1-1 LT2_DS20_S99_L001_R1_001.fastq.gz --pe1-2 LT2.highdepth/LT2_DS20_S99_L001_R2_001.fastq.gz -o /home/local/AMC/waalkes/LT20

As I increase coverage the assembly size increases dramatically. Here are the numbers.
20x - 5.3M
50x - 5.9M
100x - 7.0M
300x - 10.9
600x - 14.2M

Any help? Thanks

[asl]

I would suggest to check for possible contamination.

[waalkes]

I don't think that is the problem. Same libraries work fine with ABYSS and masurca and when I align and variant call with them.

[asl]

Well, SPAdes assembles whatever you're having in your input reads. It does not do any coverage filtering like other assemblers do. The typical source of such increase of assembly size is either contamination or some strain variability within input data. In any case, since you're having the output you could easily check what's inside.