Well said. I'll add some scientific esoterica, because it parallels software a little: we've even had custom endonuclease services for a while (TALENS[1]), which serve a very similar function. But they were hard to generate and difficult to work with. Companies like Invitrogen even sold a TALEN-making service, costing in the dozens-of-thousands of dollars to generate a TALEN for preclinical drug discovery use.
Then CRISPR came along. It was like the open-sourced, better-performing alternative to the cumbersome, proprietary software. Switching was a no-brainer, and it has handily become the future, if not the mainstream already.
Right. My own PhD work was on designing custom transcription factors to bind to specific sequences (not even with TALENs- it tried to do full molecular dynamics to predict the binding constant for multiple different DNA sequences, which was absurdly expensive). It would have to be re-engineered once for each sequence; with CRISPR, you just provide a matching template sequence.
But CRISPR/Cas9 is not open source. It is proprietary! A half billion year old natural system has been slightly tweaked and is now owned by the discovering groups. It is free for research use and for commercial purposes might still be quite expensive.
Comments
Well said. I'll add some scientific esoterica, because it parallels software a little: we've even had custom endonuclease services for a while (TALENS[1]), which serve a very similar function. But they were hard to generate and difficult to work with. Companies like Invitrogen even sold a TALEN-making service, costing in the dozens-of-thousands of dollars to generate a TALEN for preclinical drug discovery use.
Then CRISPR came along. It was like the open-sourced, better-performing alternative to the cumbersome, proprietary software. Switching was a no-brainer, and it has handily become the future, if not the mainstream already.
[1]https://en.wikipedia.org/wiki/Transcription_activator-like_e...
Right. My own PhD work was on designing custom transcription factors to bind to specific sequences (not even with TALENs- it tried to do full molecular dynamics to predict the binding constant for multiple different DNA sequences, which was absurdly expensive). It would have to be re-engineered once for each sequence; with CRISPR, you just provide a matching template sequence.
But CRISPR/Cas9 is not open source. It is proprietary! A half billion year old natural system has been slightly tweaked and is now owned by the discovering groups. It is free for research use and for commercial purposes might still be quite expensive.