Frameshift

Although they were sheltered from the wind here by the low wall, his arm moved back and forth as if blown by a breeze only it could feel. “Blame it,” said Pierre, “on my Huntington’s.”

Avi’s eyes narrowed and then he nodded, turned, and leaned back against the wall, exhausted not just by the climb but also by years of chasing Ivans and Adolphs and Heinrichs. He closed his eyes and exhaled slowly, waiting for the medics to arrive.

Chapter 42

As soon as visiting hours began, Molly came into Pierre’s room at San Francisco General Hospital. Pierre looked up at her from the bed. The left side of his face was bandaged, and his legs were in traction.

“Hi, honey,” said Molly.

“Hi, sweetheart,” said Pierre. He gestured at all the equipment hooked up to him. “After you left yesterday, somebody said my total hospital bill is going to be in the neighborhood of two hundred thousand dollars.” He managed a grin. “I’m sure glad Tiffany talked me into the Gold Plan.”

“I brought you a newspaper,” said Molly, pulling a copy of the San Francisco Chronicle out of the canvas bag she was carrying.

“Thanks, but I don’t feel much like reading.”

Molly said, “Then let me read it to you. There’s a front-page story by that man we met, Barnaby Lincoln.”

“Really?”

“Uh-huh.” She cleared her throat. “ ‘Officials from the California State Insurance Board, escorted by eight state troopers, today seized control of Condor Health Insurance, Inc., of San Francisco, in the wake of startling revelations made last week. ”Condor is out of business, as of today,“ said Clark Finchurst, State Insurance Commissioner. ”The industry’s emergency fund, which was established to handle such things, will take care of current claims until Condor’s policies can be handed over in an orderly fashion to other insurers.“ ’ ”

“All right!” said Pierre.

“It says there’s going to be a full inquiry. Craig Bullen is cooperating with the authorities.”

“Good for him.”

“Oh, and I picked up that printout you wanted.” She took a two-inch-thick pile of fanfold computer paper out of her bag and placed it on the table beside his bed.

“Thanks,” said Pierre.

Molly sat down on the edge of the bed and took one of Pierre’s dancing hands in hers. “I love you,” she said.

“And I love you, too,” said Pierre, squeezing her hand. “I love you more than words can say.”

Pierre lay in his hospital bed that night. His six minutes of CPU time on LBNL’s Cray supercomputer had at last become available, and the simulation he and Shari had coded had finally been run. Pierre started wading through the 384 pages of printout.

When he was done, he operated the hand control that lowered the motorized back of his bed. He stared at the ceiling.

It made sense. It all fit.

The existence of codon synonyms did indeed allow additional information to be superimposed on the standard A, C, G, T genetic code.

Yes, AAA and AAG both made lysine, but the AAA form also coded a zero into what Shari had already dubbed, in a note jotted in the margin, “the gatekeeper function,” which governed the correction or invocation of frameshift mutations. Meanwhile, the AAG version coded a one.

But that was just the tip of the iceberg. There were four valid codons that made proline: CCA, CCC, CCG, and CCT. For these, the final letter indicated a base-sixteen order of magnitude shift of the splicing cursor, which marked the position where a nucleotide would be added or deleted from the DNA, causing a frameshift. The CCT form moved the cursor sixteen nucleotides; the CCC form moved it 16, or 256 nucleotides; the 2

CCA form 16, or 4,096 nucleotides; and the CCG form moved it l6, or 3 4

65,536 nucleotides.

Other synonyms performed different jobs: GAA and GAG both made glutamine, but they also set the direction of the splicing cursor’s movement. GAG set it moving to the “left” (in the direction leading from the three-prime carbon to the five-prime carbon in each deoxyribose), and GAA set it moving to the “right” (the five-prime to three-prime direction).

Meanwhile, TTT, which made phenylalanine, coded for a nucleotide insertion, while its synonym TTC was the instruction for a nucleotide deletion. And the four codons that made threonine — ACA, ACC, ACG, and ACT — indicated by their final letter which nucleotide would be inserted at the splicing cursor.

The coding based on synonyms moved the cursor, but the timing of when frameshifts would be invoked was governed by certain of the seemingly endless stuttering sequences in the junk DNA. On the smaller scale of the individual, it had already been demonstrated that the number of CAG stutters set the age at which Huntington’s would first manifest itself, and, as Pierre had pointed out to Molly, the number of repeats does change from generation to generation in a phenomenon called “anticipation” — an ironically prophetic name given what Pierre and Shari’s model showed.