How much more HTS do we need to do

Several colleagues have brought to my attention the recent announcement that Japan is joining global efforts to fight neglected diseases through a public-private partnership.I am excited because Japan is a major scientific power with many pharmaceutical companies that could offer up their compounds and scientists to come up with new avenues to treat diseases like tuberculosis.

I think we can help them to not repeat what has already been done, as I noticed the statement at the outset “Tens of thousands of new compounds will be screened for effectiveness against critical illnesses”. I just hope this is not going to be a random screening exercise. By my back of the envelope calculation, many millions of compounds have already been screened by Pharma (GSK alone have done about 2 million with just a tiny fraction of the data publically available for TB ), AZ, Novartis etc have probably screened 100’s of thousands,  NIH funded screening adds a similar number with over 300,000 compounds and the latter data is in the public domain. For malaria, MMV collaborating groups are probably close to have screened 10 million compounds at a guess.

Which begs the question…we clearly now have enough (more than enough) data to understand what are the important properties to enable a molecule to have whole cell activity versus tuberculosis, and we  have recently shown how predictive the computational models can be learning from just several thousand high quality data points here and here..with extensive prospective testing.

So my thought is this, the companies and groups involved can use the models to pick the compounds worth testing, saving themselves time and money. We are happy to make them freely involved to anyone involved.. just contact me. I may even just put the models on Figshare!

P.S. probably what we need to do now is create models for compounds that are active in in vivo models of the disease.

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>