constructing de novo assembly of plant short reads

Question

[analyst]

Dear all,

I have Illumina short reads (paired-end) plant data for which reference genome is not available. I have to construct assembly with short reads only as I do not have long reads. Plant contains tetraploid and diploid varieties I identified best Kmer 37 and 27 respectively through kmergenie without --diploid parameter as it was omitting some information. Please note that read length is 80-127bp, sequencing depth is 14.9X, total sequences are 4-5 million. Total samples are 40, out of 40 paired end samples 20 are diploid varieties and 20 are tetraploid varieties. Do I need to construct two genomes w.r.t. ploidy?

Please guide me if I have used correct approach. Also which short reads assemblers should I use for plant data.

I tried AbySS tool for multiple samples together using command such as:

abyss-pe np=60 k=37 name=FA2 B=1G in='FA2_18_1.fastq.gz FA2_18_2.fastq.gz FA2_19_1.fastq.gz FA2_19_2.fastq.gz FA2_20_1.fastq.gz FA2_20_2.fastq.gz FA2_21_1.fastq.gz FA2_21_2.fastq.gz FA2_22_1.fastq.gz FA2_22_2.fastq.gz' 

Can you please confirm if I can assemble multiple samples together.

Best regards,

Bushra

[analyst]

Abyss output scaffolds.fa file contains short kmer sized scaffolds. I am putting first few lines of scaffolds.fa file here as:

>0 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
CCAGAGCATCTACTAGCAACGGAGAGCATGCAAGATC
>2 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
AGAGCATCTACTAGCAACGGAGAGCATGCAAGATCAC
>4 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
AGCATCTACTAGCAACGGAGAGCATGCAAGATCACAA
>8 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
TCTACTAGCAACGGAGAGCATGCAAGATCACAAATAA
>9 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
CTACTAGCAACGGAGAGCATGCAAGATCACAAATAAC
>11 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
ACTAGCAACGGAGAGCATGCAAGATCACAAATAACAT
>17 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
AACGGAGAGCATGCAAGATCACAAATAACATATGATA
>21 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
GAGAGCATGCAAGATCACAAATAACATATGATAAATA
>27 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
ATGCAAGATCACAAATAACATATGATAAATAAATAAT
>31 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
AAGATCACAAATAACATATGATAAATAAATAATTGAT
>32 37 255 read:bn,LH00330:195:227VG7LT4:6:1103:0:451604/1
AGATCACAAATAACATATGATAAATAAATAATTGATC
>36 37 255 read:bn,LH00330:195:227VG7LT4:6:1117:0:5202810/1
CCATCGAGGTATCCCCTACGACCAACTCCAAATATAG

My concern is that why scaffolds are too short of kmer size that is 37. In abyss kmers are not combined to make up contigs and then scaffolds ?

Answer 1

[Dave Carlson]

You could try Spades or MaSuRCA, but between tetraploidy and the general tendency of (most) plants to have highly repetitive genomes, a short read only assembly is very likely to be extremely fragmented and not as useful as you'll want it to be.

[analyst]

Thankyou so much Dave. Is not spades particularly designed for bacterial genomes?

What do you suggest about abyss, velvet etc.

Thanks alot

[Dave Carlson]

I think Spades was originally designed for bacterial genomes, but it can be used for eukaryotes as well, as far as I recall. That said, I've only ever used it to assemble bacteria genomes.

Abyss is another good option. I don't think Velvet has been updated in quite some time, so I would not be surprised if its performance is not as good on modern hardware compared to other tools, but that's just speculation.

Overall, though, whichever tool you use, I fear that your genome assembly will be less than satisfactory due to the reasons mentioned previously.

[analyst]

Thanks for your valuable input!

[analyst]

I am going to give both suggested tools a try. Thankyou Dave.

Can I assemble multiple paired end reads into one assembly using these tools or will I have to generate assemblies for each sample separately?

Thankyou for your help!

[Dave Carlson]

If you have multiple samples from the same species, you can just combine the fastq files and generate a single assembly.

[analyst]

Do you mean merging of fastq files and running assembly on merged fastq file?

For example for MaSuRCA I found following command that seems for one sample only how can I adjust it for multiple samples paired end reads to get single assembly.

/path_to_MaSuRCA/bin/masurca -t 32 -i /path_to/pe_R1.fa,/path_to/pe_R2.fa

[Dave Carlson]

You can just concatenate the various R1 and R2 fastq files before running the assembly. Let's say you have two samples (sample1, sample2), then you could do the following:

cat sample1_R1.fastq sample2_R1.fastq > combined_R1.fastq
cat sample1_R2.fastq sample2_R2.fastq > combined_R2.fastq

And then run the assembly using the combined_R1.fastq and combined_R2.fastq files.

[analyst]

Great, thanks!