This can happen for highly fragmented assemblies or poor quality libraries. You can specify library statistics manually.
To do so, look for *.is.txt file specific for your library i.e. for -i 5000_1.fq.gz 5000_2.fq.gz you will have to enter requested values into 5000_2.fq.gz.is.txt.
.is.txt file is tab-delimited, with following columns:
- read length
- median insert size
- mean insert size
- standard deviation of insert size
- and number of reads with FF, FR, RF & RR orientation.
Make sure you specify some large number of mates for requested orientation (ie 100,000). For example if you want mate-pairs with RF orientation and 5kb insert size +/- 1.5kb, enter into respective *.is.txt file:
60 5000.0 5000.0 1500.0 0 0 100000 0Redundans removes all contigs, but the longest one, that fullfill identity & overlap critaria during reduction step. For more info see issue #8.
https://github.com/Gabaldonlab/redundans is the official repository for Redundans, but we keep https://github.com/lpryszcz/redundans for back-compatibility, as some of the very first users of Redundans use it. Both Redundans repositories contain the same code and are regularly updated.
Leszek P. Pryszcz and Toni Gabaldón (2016) Redundans: an assembly pipeline for highly heterozygous genomes. NAR. doi: 10.1093/nar/gkw294