Release date: 2016-09-14
Researchers at Stanford University are designing (new) methods that allow computers to help perform intensive genetic analysis that scientists need to do manually when studying patient genomes to diagnose disease.
Shayla Haddock
In 1997, when Shayla Haddock was born, her parents immediately realized that something was wrong. As the sixth child in the family (seven children), Shayla's facial features are different, the feet are deformed, and the limbs are shorter than normal, and the body shape is smaller than most newborns. Hearing tests indicate that she still has deafness problems.
In 2012, the doctor analyzed Shayla Haddock's genome for reasons that caused his foot deformity, deafness, and limbs to be shorter than normal. Although the initial analysis did not provide a diagnosis, a reanalysis in 2015 revealed the answer: Shayla suffers from a rare genetic disease called Wiedemann-Steiner syndrome. This reanalysis used the newly developed computing tools from Stanford University.
Shayla's parents, Cheryl and Levko Siloti, looked for these unusual answers. They were worried about whether Shayla's symptoms were caused by some preventable problems during Cheryl's pregnancy. Can you diagnose and improve Shayla's treatment plan? If Shayla’s brothers and sisters are married in the future, will their children face the same risk of disease?
“This is an emotional roller coaster,†says Cheryl Siloti. Over the years, doctors have made many possible diagnoses about Shayla's condition, but medical examinations have repeatedly denied their theory. “The doctor told us about these possibilities, and then they denied that these might be 'no, this is not the answer'â€.
The problems faced by this family in Stockton, Calif., illustrate the challenge of diagnosing rare genetic diseases and an example of why scientists at the Stanford University School of Medicine designed new methods and how to help.
Doctors are as eager to make a diagnosis as Shayla's parents, but they can hardly produce any definitive results. August 10, 2012 - Doctors at the Lucile Packard Children's Hospital Stanford concluded that "the genetic patterns and symptoms of Shayla cannot be matched to (any) a disease" Conclusion Two weeks later - a scientific report revealing new findings about the link between genetic defects and a rare disease, this article allows doctors to make a diagnosis of Shayla. But at that time, it was not possible to periodically reanalyze the results of genetic testing to take into account new knowledge. Shayla's family and doctors still don't know they can get an answer.
——Genetic reanalysis
Last year, as part of a scientific study, Shayla's parents agreed to reanalyze her genome. This time, computer scientists at Stanford University used the new computational tools they developed to compare Shayla's gene sequences with scientific literature. They discovered the 2012 (the) scientific report and predicted that Shayla was suffering from a rare genetic disease called Wiedemann-Steiner syndrome, which was confirmed by Dr. Shayla.
“Every month, more global genetic diversity is reflected in the scientific database, and every time there is more information, it’s easier to explain what you see next.†Shayla’s clinical experience at Packard Children’s Hospital Geneticist Dr. Jon Bernstein said. Dr. Jon Bernstein is the author of a new report published online July 21 in Genetics in Medicine. Ten percent of the patients in the study—including Shayla—have not been diagnosed after the first genetic analysis (on average 20 months ago) – diagnosed as a different rare disease based on recent findings.
These “missing omissions†highlight a huge challenge in the field of precision health: although the speed, cost and workload of acquiring individual gene sequences have declined dramatically in recent years, well-trained experts still need to work about 20 to 40 Hours to match a patient's rare mutations with scientific literature information that may reveal the diagnosis. Of the patients suspected of having a rare genetic disease, 75% were unable to obtain a diagnosis after the first DNA analysis. However, the knowledge base is growing rapidly, and researchers find out about 250 causes of genetic disease each year, while finding the association between 9,200 specific gene causes and known diseases.
- too much requires manual diagnosis
"Our research shows that reanalysis of patient genetic test results is useful because of the steady rate of discovery," Bernstein said. Bernstein is also an associate professor of pediatrics at Stanford University School of Medicine.
“But we don’t have enough human resources to continue all manual analysis as clinicians and scientists have done in the past,†said Dr. Gill Bejerano, associate professor of developmental biology, computer science and pediatrics.
Computer scientists led by Bejerano have designed automated methods used in new research. Millions of Americans may suffer from some form of rare genetic disease, he points out – too much for manual diagnosis. “Our team believes that it is more meaningful to build time to build computer science tools that can help us do a lot of work, compared to tens of hours of continuous analysis in each patient,†he said.
In the new study, these scientists tested whether automated comparisons between undiagnosed patient genomes and existing gene databases could speed diagnosis. The results show that this method is effective.
“The genome is basically a programming language,†says Bejerano. “We really want to use machine learning and other methods to build computer systems and leave as little work as possible to human experts. In these jobs, computers are not as good as humans, but we I think we can handle 80% to 90% of the work in this method by computer and save a lot of time for humans during the cycle."
- Compare the genes of patients and parents
According to Bernstein and Bejerano, another key finding of the new study is that comparing the patient's genetic sequence to the parent's genetic sequence can greatly speed up the diagnosis process. This contrast helps to discover new pathogenic mutations that occur in patients without their parents. "If (patient) parents' data is in front of you, these (discovery) will be easier to visualize," Bernstein said.
In Shayla's case, her diagnosis brought the family the answers they had been seeking. Her pathogenic mutations are not inherited from parents, but occur naturally. This is not a preventable problem, nor will it affect the children of her brothers and sisters. "Knowing this, it really made us feel a long sigh of relief," Siloti said.
The diagnosis also helped Silotis find other families with the same children. They share stories in Facebook groups and feel that they have discovered the new meaning of support and group. “We always believe that knowledge is power,†Siloti said. “It feels good to get the answer, especially after a long search.â€
The new study was funded by the Department of Pediatrics at Stanford University, the Stanford Discovery Fund, the Defense Advanced Research Projects Agency, and the National Institutes of Health (U01MH105949). Co-leaders are research scientists Dr. Aaron Wenger and Dr. Harendra Guturu.
Why it is difficult to diagnose rare genetic diseases, the following data can bring inspiration:
1. The size of the human genome: 3 billion bases. A base is a single "letter" in the "word" that makes up our gene and the genomic region of the control gene;
2. The number of Americans suffering from a rare form of genetic disease is estimated to be: 25 million;
3. The number of single-gene diseases that newly identify the cause each year: 250;
4. The number of newly identified genetic alterations associated with existing diseases each year: 9200;
5. Time required to analyze individual genetic information when suspected of a rare genetic disease: 20 to 40 hours;
6. Proportion of the first genetic sequencing in patients with suspected rare monogenic diseases to obtain a diagnosis: 25%;
7. Typical frequencies for reanalysis to help undiagnosed patients match new findings: have been rare, but are expected to become more common thanks to new automated genetic analysis techniques.
Source: China Rare Disease Network
Qingdao Beautiful Skin Biotechnology Co., Ltd , https://www.hafilleresthetic.com