Can I use both paired and single end reads in the same MEGAHIT run?

Question

[dark.lord]

Hi everyone,

I have this doubt - can I use both paired and single end reads in the same MEGAHIT run? I launched this command:

megahit --preset meta -1 1_trim4p_R1.fastq.gz,2_trim4p_R1.fastq.gz,3_trim4p_R1.fastq.gz,4_trim4p_R1.fastq.gz,5_trim4p_R1.fastq.gz,6_trim4p_R1.fastq.gz -2 1_trim4p_R2.fastq.gz,2_trim4p_R2.fastq.gz,3_trim4p_R2.fastq.gz,4_trim4p_R2.fastq.gz,5_trim4p_R2.fastq.gz,6_trim4p_R2.fastq.gz –r 1_trim4s_R1.fastq.gz,2_trim4s_R1.fastq.gz,3_trim4s_R1.fastq.gz,4_trim4s_R1.fastq.gz,5_trim4s_R1.fastq.gz,6_trim4s_R1.fastq.gz,1_trim4s_R2.fastq.gz,2_trim4s_R2.fastq.gz,3_trim4s_R2.fastq.gz,4_trim4s_R2.fastq.gz,5_trim4s_R2.fastq.gz,6_trim4s_R2.fastq.gz -m 0.25 -t 15 -o outputPS/

Where in -1 and in -2 are the trimmed (Trimmomatic) paired end reads, whereas in -r all the unpaired/single end reads (as well from Trimmomatic).

The assembler ran, but then checking the log it seems it didn't include the single reads in the run. Log:

MEGAHIT v1.0.3
--- [Mon Feb 26 19:12:05 2018] Start assembly. Number of CPU threads 40 ---
--- [Mon Feb 26 19:12:05 2018] k list: 21,41,61,81,99 ---
--- [Mon Feb 26 19:12:05 2018] Converting reads to binaries ---
/opt/bio/megahit-1.0.3/megahit_asm_core buildlib ./megahit_out/tmp/reads.lib ./megahit_out/tmp/reads.lib
    [read_lib_functions-inl.h  : 209]     Lib 0 (1_trim4p_R1.fastq.gz,1_trim4p_R2.fastq.gz): pe, 5262952 reads, 285 max length
    [read_lib_functions-inl.h  : 209]     Lib 1 (2_trim4p_R1.fastq.gz,2_trim4p_R2.fastq.gz): pe, 4700864 reads, 285 max length
    [read_lib_functions-inl.h  : 209]     Lib 2 (3_trim4p_R1.fastq.gz,3_trim4p_R2.fastq.gz): pe, 5187702 reads, 285 max length
    [read_lib_functions-inl.h  : 209]     Lib 3 (4_trim4p_R1.fastq.gz,4_trim4p_R2.fastq.gz): pe, 4999452 reads, 285 max length
    [read_lib_functions-inl.h  : 209]     Lib 4 (5_trim4p_R1.fastq.gz,5_trim4p_R2.fastq.gz): pe, 6113712 reads, 285 max length
    [read_lib_functions-inl.h  : 209]     Lib 5 (6_trim4p_R1.fastq.gz,6_trim4p_R2.fastq.gz): pe, 4047104 reads, 285 max length
    [utils.h                   : 124]     Real: 64.6702 user: 45.7480   sys: 4.2520 maxrss: 162880

Whereas the opts.txt file, which reports the command launched, included (of course) the single reads after -r:

--preset
meta
-1
1_trim4p_R1.fastq.gz,2_trim4p_R1.fastq.gz,3_trim4p_R1.fastq.gz,4_trim4p_R1.fastq.gz,5_trim4p_R1.fastq.gz,6_trim4p_R1.fastq.gz
-2
1_trim4p_R2.fastq.gz,2_trim4p_R2.fastq.gz,3_trim4p_R2.fastq.gz,4_trim4p_R2.fastq.gz,5_trim4p_R2.fastq.gz,6_trim4p_R2.fastq.gz
–r
1_trim4s_R1.fastq.gz,2_trim4s_R1.fastq.gz,3_trim4s_R1.fastq.gz,4_trim4s_R1.fastq.gz,5_trim4s_R1.fastq.gz,6_trim4s_R1.fastq.gz,1_trim4s_R2.fastq.gz,2_trim4s_R2.fastq.gz,3_trim4s_R2.fastq.gz,4_trim4s_R2.fastq.gz,5_trim4s_R2.fastq.gz,6_trim4s_R2.fastq.gz
-m
0.25
-t
15
-o
outputPS/

Am I doing something wrong? Is there any sort of conflict using both paired and single ends in the same run?

Thanks everyone.

Answer 1

[dark.lord]

Update: found a mistake in the command: it's -r, not –r. It does work! The difference is not visible but these are two different ASCII code characters.