I have assembled contigs from fastq reads of sequenced phage data. The best contig size is of 88000bp with k-mer coverage of 49. Now I am trying to get the consensus by mapping the reads back to this contig. The question is , should I do this step to get the consensus or SPAdes resultant contig is a consensus itself and I can use it for gene prediction and annotation? Please let me know if anything is not clear.