Paired end reads in MALT and MEGAN


I have a mix of both single-end reads and unmerged paired-end reads per sample. I have separated these reads into two different FASTQ files (single and paired end), but have some questions regarding the next stages of my analysis:

  1. I am using MALT for alignment. Does MALT need to be ‘told’ that some files consist of paired end reads?

  2. When I import the .rma6 MALT output into MEGAN, how do I inform MEGAN that reads in a file are paired-end? Is there any way to subsequently analyse the alignments from the paired-end and single-end alignments per sample as one sample?

Many thanks!

MALT doesn’t have a paired-read mode, so it won’t attempt to analyze reads in pairs.
To do paired read processing, set MALT to generate SAM files and then import the SAM files into MEGAN, specifying paired read mode there. If you have multiple SAM files for the same sample, then import them all at the same time to create one unified rma6 file.

Dear Daniel,
I tried this but I failed. The “-f SAM” or “–format SAM” does not seem to work, pls see the error message below. I want to Compare several samples rather than twice several “half-samples”, that’s why I’d like to combine/join/merge the two directions each that I preprocess using MALT. With no “-f” option the whole thing work nicely, btw.

--------- error message (output) follows -------------
malt-run -i NR05_S21_R1_trimG.fastq NR05_S21_R2_trimG.fastq --format SAM -d /mnt/sdd1/data/databases/malt/Silva_tax_index/ -m BlastN -o . -t 24
Version MALT (version 0.4.0, built 6 Sep 2017)
Copyright © 2017 Daniel H. Huson. This program comes with ABSOLUTELY NO WARRANTY.
jloda.util.UsageException: Invalid, unknown or duplicate option: --format SAM
, use option ‘-h’ for help
at jloda.util.ArgsOptions.done(
at malt.MaltRun.main(
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(
at sun.reflect.DelegatingMethodAccessorImpl.invoke(
at java.lang.reflect.Method.invoke(
at com.exe4j.runtime.LauncherEngine.launch(
at com.install4j.runtime.launcher.UnixLauncher.main(