I built several de novo transcriptomes from non-model organism without genome reference by Trinity. Sequencing libraries were prepared with stranded method (dUTP), therefore I only used paired-end reads with RF SS_lib_type parameter in Trinity for assembling. However, earlier preprocessing steps retained ~20% amount of data as singletons (reads that lost their pair due to filtering), I really want to use this data as single-end reads to simultaneously map to de novo transcriptome together with PE reads. My final purpose is performing Differential expression analysis of coding mRNA.
Could you please suggest any tools for mapping both SE and PE reads to transcriptome, is BWA a good choice? Furthermore, does strand information have any benefit in case of mapping reads on transcriptome? (since we do not know reference genome sequence)
Thank you very much in advance !