Lynn Whittaker stood in the hallway of her home looking at the framed photos on the wall. In one, her son, Andrew, is playing high school water polo. In another, he's holding a trombone.

The images show no hint of his life today: the seizures that leave him temporarily paralyzed, the weakness that makes him fall over, his labored speech, his scrambled thoughts.

Andrew, 28, can no longer feed himself or walk on his own. The past nine years have been a blur of doctor appointments, hospital visits and medical tests that have not produced answers.

"You name it, he doesn't have it," his mother said. Andrew has never had a clear diagnosis. He and his family are in a torturous state of suspense, hanging their hopes on every new exam and evaluation.

Recently, they have sought help from the Undiagnosed Diseases Network, a federally funded coalition of universities, clinicians, hospitals and researchers dedicated to solving the nation's toughest medical mysteries. The doctors and scientists in the network harness advances in genetic science to identify rare, sometimes unknown, illnesses.

The Undiagnosed Diseases Network was founded in 2015 with a $43 million grant from the National Institutes of Health. Building on work already being done at NIH, the initiative expanded to include universities across the country: Duke, Columbia and Stanford are among the other sites. The goals are to provide answers for patients with mysterious diseases and to learn more about the disorders.

Since its launch, the network has received nearly 1,400 applications on behalf of patients. It has accepted 545 for review so far. Just 74 of the cases have been diagnosed, including 11 at UCLA. Andrew Whittaker's case is among many in progress.

It's like battling "an unknown enemy," said Euan Ashley, one of the principal investigators of the network's Stanford University site. "That is a particular form of torment that other patients don't have."

Researchers throughout the network use advanced medical technology. For example, to study patients' gene expression and disease progression, they can make models using nearly transparent zebrafish, whose genetic structure is similar to that of humans. And scientists can conduct whole genome sequencing, which allows the medical team to read a patient's DNA and identify changes that can reveal what may be causing a disease.

"We have powerful techniques to look at every gene that is being expressed as well as every gene that is inherited," said Stanley Nelson, one of UCLA's principal investigators and the lead doctor on Andrew's case. "This is an example of true precision medicine."

Nelson said the network can examine all known genes — not just the ones believed to have mutations that cause diseases. Doing that can lead to the discovery of new illnesses.

"Part of what we have to do is keep building that library, that encyclopedia of what gene and what gene mutations cause what symptoms," Nelson said. "It's just incomplete at this moment."

Doctors at one institution might think their patient is a unique case, only to learn that colleagues elsewhere have a patient with a similar illness. But even when diseases are diagnosed or gene mutations are discovered, treatments may still not be available.

Andrew Whittaker's odyssey began one afternoon at age 19, when he started trembling and couldn't speak. Doctors suspected he was suffering from anxiety and prescribed medication to control it. But Andrew said he continued to have "episodes," during which everything just went blank.

Andrew also started losing his balance and falling off his bicycle. The family visited several hospitals. Doctors found that the receptors in his brain were malfunctioning and that he lacked sufficient dopamine, a chemical compound in the body responsible for transmitting signals between nerve cells. As a result, Andrew has some symptoms similar to those of Parkinson's disease. Still, Andrew's symptoms didn't add up to any known disease.

After a week of tests with the Undiagnosed Diseases Network, Nelson sat down with the family to explain what he'd found. He had reviewed Andrew's genome and compared it with that of both parents. Andrew had one copy of a defective gene that leads to Parkinson's but the genome sequencing didn't show a second copy, without which it could not be Parkinson's.

He explained that Andrew's illness was clearly progressive and that his brain was shrinking, making it harder for him to process language and information. Nelson said he still didn't have a diagnosis — he believed it was a brand-new disease.

Lynn Whittaker was disappointed. "We are still left with just hope that they will come up with something," she lamented. "What else do we have?"

Andrew said his relatives have asked if he's afraid the doctors will find something. "I'm more scared if they don't," he said.