A company is offering 1Gb for $xxx. Since it is supposed that 1Gb=10^9 bases, is it enough for assemble a whole bacterial genome (~9Mb)?
Without knowing the specifics of the sequencing like library prep method, sequencing technology, read length, and paired end distance (if paired end sequencing is done) and the organism, this is hard to predict.
I agree that 100x coverage should give you most to all of the sequence in most scenarios. If we are talking about something with high G+C and Illumina, you want to get the TruSeq PCR-free library (and avoid Nextera XT!).
Beyond that, it depends if you want to really get the genome completely closed (or at least down to one scaffold per replicon). For that you would need MatePair data with a paired size of >6kb (to span over the RRNs). Otherwise the reasons given by Xavier will leave you with scaffolds interrupted by repetitive elements.
I agree with Xavier, if they offer 100X coverage using PE I would go ahead. You may map it later using reference genomes if its the case and in the worse cases you can always try the novo-Assembly.
I agree with the thread above, but I would add that we've been finding that the short read data can be complemented with even just a little bit of long-read (PacBio) data which is quite inexpensive. For about $100 of long-read sequence data, you an go from 10s of contigs on microbial genomes down to a single or few contigs.
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis
- genetic techniques
- Sequencing