Ten, 20, 30, 50 years from now.
I would imagine plenty of data ready to analyse, with this old fashioned '-omics' concepts almost forgotten ;) , possibly due to data integration, both at wet and in silico levels. Also, lab procedures to get data being quite fast.
What about you guys?
As data accumulates, people realize that much of the generated sequence data is unusable due to noise/lack of precise metadata. So I think one trend is that data will be splintered into smaller manually curated sets, many of which will either require a license or will simply not be available to public at all. As an example, GenBank is infested by lack of precise/correct metadata concerning e.g. taxonomy. I think in 20 years we will certainly have very long and accurate reads, so the amount of annually generated data may actually decrease dramatically and the problems that algorithms have to solve then will often be less difficult than they're now. Hard to say how things are 50 years from now. My guess is that Linux will still be around. It will probably outlive us all and even guide space ships to other planets, some day. That is unless systemd takes over the kernel too..
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
10 , 20, 30, 50 years from now: everyone is looking for some new antibiotics, but that's too late.
We may have found some new ones on Mars.
yes and the mars mission also brought back a new viral strain that kills the population that fled inland because of the rising sea levels ...
Current analysis will have become highly automated leading to the reduced need for bioinformatics specialists, the community responds by developing increasingly complex forms of analysis to remain employed.
Illumina continues to soak up IP to keep people buying their HiSeq 1000000 machines.
Oxford Nanopore still hasn't released their nanopore sequencers to the public.
NCBI still provides poor documentation for all their tools.
Someone invents SuperMassive Data, the next step after Big Data. Continues to convince managers/etc that it is a totally new way to deal with the huge piles of data. The market becomes flooded with hundreds of tools doing the same things that totally aren't reinventing the wheel for no real reason.
PacBio introduces their RLY-SMRT sequencing, capable of 500Mb reads with 30% error rate.
People are still doing RACE to sequence the termini of viral genomes.
(no) Thanks for dashing hopes of many :-)
Quite frankly. 20 years ago no one could imagine where bioinformatics is today, so I don't think it's possible to imagine where we are in 20 years from now. But I am sure it will be still relevant. :p
In 1953 DNA-sequence principle was discovered by Crick.
Some years later Sanger-sequencing technique was suggested.
In 2001 to sequence a small bacteriphage took from several months to half a year,
but there was almost no mistake in the sequence.
In 2006-2010 scientists learned about human genome. but it is still a mistery.
Now NGS-approaches are 100-1000 times cheaper than even 20 years ago, but the quality of such sequences
is still rather far from Sanger results...
I don't think that is true.
It's true if you consider a single Sanger read versus a single NGS read. However, this is why no use uses 1X coverage, so it's not a fair comparison. At 10X or 100X, which is a more realistic scenario, it's a different story.
Do we have numbers for this, comparing, say longer read NGS to Sanger?
I know they preferred Sanger for clinical targeted/gene-specific sequencing in my previous lab, does any other clinical lab use different techniques?
That may partly be because FDA has not approved NGS (exception: MiSeq DX for CF and some others) for wide-spread diagnostic use.
Thank you, genomax2!
Lots of places use NGS for clinical applications. For example, Foundation Medicine has been doing it for years and built a large company based on that.
I'm still writing my PhD thesis.
When should we expect it to be done? Another 10, 20 or 50 years? Hurry up already, we are holding a spot for you on the Mars bioinformatics mission :-)
DNA alignment is still problematic. There are still multi-million dollar efforts to GWAS anything.
Hey, someone was talking about this 6y ago. Future Directions In Bioinformatics Future Directions In Bioinformatics