George Church

Church's doctoral work produced the method he and Gilbert called genomic sequencing. Before it, reading a gene's sequence meant first cloning the fragment in bacteria to obtain enough identical copies. Their 1984 paper read sequences directly from total mouse DNA. The whole genome was cut and chemically cleaved, the fragments were separated by size on a gel and fixed to a nylon membrane, and a short radioactive probe picked out the ladder of bands belonging to one sequence. Each band represented 3 femtograms of DNA, and the same membrane could be probed again and again for different sequences.^[2]

In December 1984 Church was one of 19 scientists at a meeting in Alta, Utah, convened by the United States Department of Energy and an international commission on environmental mutagens. The question before it was whether new DNA methods could detect mutations directly, and in particular any rise in mutations among the survivors of the atomic bombings of Hiroshima and Nagasaki and their children. The meeting's main conclusion was that the "methods were incapable of measuring mutations with sufficient sensitivity", and the discussion it began is counted among the origins of the Human Genome Project.^[15] His Wyss Institute biography credits him with helping to initiate the project that year.^[1]

His 1988 paper on multiplex sequencing replaced each sample in a sequencing run with a mixture of many. Every DNA molecule was first joined to a short identifying sequence, a "tag"; the tagged molecules were pooled and processed together, and the resulting membranes were then probed once for each tag. After 50 rounds of probing the signal had not faded.^[3] Tags of this kind, now usually called barcodes, became a standard part of high-throughput sequencing. His laboratory transferred its automated sequencing methods and software to Genome Therapeutics, a company in Waltham, Massachusetts, which announced in December 1994 that it had sequenced the genome of the stomach bacterium Helicobacter pylori; his laboratory's site lists it as the first genome sequence, made with methods he developed.^[16][14] The first published complete genome of a free-living organism, that of Haemophilus influenzae, appeared the following year.^[17]

In 2005 his group described polony sequencing, in which millions of DNA fragments, each copied onto its own bead, were held in a gel and read cycle by cycle on what the paper calls "a commonly available, inexpensive epifluorescence microscope". It resequenced an evolved strain of Escherichia coli with fewer than one error per million consensus bases, at roughly one ninth of the cost per base of conventional sequencing.^[18] An instrument built on the method, the Polonator, topped The Scientist's list of the year's innovations in 2008.^[8]

On March 17, 1995, Church, David Deamer, Daniel Branton, Richard Baldarelli and John Kasianowicz filed a patent application titled "Characterization of individual polymer molecules based on monomer-interface interactions". It describes characterizing a linear polymer, such as a strand of DNA, "by measuring physical changes across an interface" as the molecule moves through it and its units interact with the boundary. It was granted on August 18, 1998, as US 5,795,782, and assigned to Harvard and the University of California.^[4] The following year Kasianowicz, Branton, Deamer and a fourth colleague reported single strands of RNA and DNA driven one at a time through the protein pore alpha-hemolysin, each detected by the dip it caused in the ionic current. Church is not an author of that paper, which is usually taken as the experimental start of the field.^[19]

He returned to nanopores two decades later. In 2016 he and Kasianowicz were among the 32 authors of a paper describing sequencing by synthesis read electronically on an array of nanopores. A polymerase sat at the mouth of each pore, and each of the four kinds of nucleotide carried a different polymer tag; as the enzyme built a strand, the tag of each incoming nucleotide entered the pore and produced its own level of current blockade.^[20]

In 2005 Church started the Personal Genome Project, which recruits volunteers who agree to make their genome sequences and trait information public.^[5] Its first ten participants, known as the PGP-10, included Church himself and the technology writer Esther Dyson; they received the first data released by the project in October 2008.^[21] The nonprofit that runs it, PersonalGenomes.org, which Church directs, describes itself as providing "the world's only open-access information on human Genomic, Environmental & Trait data".^[14]

In 2009 his laboratory described multiplex automated genome engineering (MAGE), which edits many sites on a bacterial chromosome at once. Applied to the pathway by which E. coli makes isoprenoids, it changed 24 genetic components simultaneously, generated "over 4.3 billion combinatorial genomic variants per day", and produced strains making more than five times as much of the red pigment lycopene within three days.^[22] Four years later a team including his group replaced every known UAG stop codon in an E. coli genome with the synonymous UAA, deleted the protein that reads UAG as "stop", and so freed the codon for new uses. The resulting genomically recoded organism incorporated nonstandard amino acids more readily and resisted infection by the bacteriophage T7.^[23]

In January 2013 his laboratory reported the use of the bacterial CRISPR system with custom guide RNAs to edit human cells. The paper appeared in Science alongside a parallel report from Feng Zhang's laboratory at the Broad Institute.^[24][25] In 2014 Church and colleagues described how CRISPR-based gene drives might spread an engineered trait through a wild population, proposed safeguards, and called for "thoughtful, inclusive, and well-informed public discussions" before any use.^[26] In 2015 his group inactivated all 62 copies of the porcine endogenous retrovirus in a pig cell line, the copy count it had first had to measure, and reported a more than thousandfold drop in transmission of the virus to human cells, a step toward transplanting pig organs into people.^[27]

In 2012 Church, Yuan Gao and Sriram Kosuri "wrote a 5.27-megabit book using DNA microchips, and read the book by using next-generation DNA sequencing".^[6] The book was Regenesis: How Synthetic Biology Will Reinvent Nature and Ourselves, which Church wrote with Ed Regis. Harvard Medical School noted that before it reached the shops on October 2, 2012, its author had already made 70 billion copies, "roughly triple the sum of the top 100 books of all time", at a density of "5.5 petabits, or 1 million gigabits, per cubic millimeter".^[28][29]

In 2017 his group used the CRISPR system's own memory, the Cas1 and Cas2 proteins with which bacteria file away fragments of invading viruses, to write the pixel values of black-and-white images and a short movie into the genomes of a population of living bacteria.^[7] The movie was five frames of a galloping mare from Eadweard Muybridge's Animal Locomotion (1887), added to the cells one frame a day.^[30]

In 2012 Church was one of six authors of "The Brain Activity Map Project and the Challenge of Functional Connectomics", a proposal to record the activity of every neuron in a circuit.^[31] The proposal was a starting point for the United States BRAIN Initiative, which his laboratory site lists among the projects he co-initiated.^[14]