After graduating from Yale  University and earning her medical degree at Harvard Medical School , Shafali Jeste did a child neurology residency at Children's Hospital Boston . Toward the end of that residency, she realized she wanted to do research focused on autism. "I just saw these kids, and I was fascinated by them," she says of her young autistic patients. "I couldn't believe that we didn't understand what was going on in their brains to make them work like that."
So Jeste designed a clinical fellowship in behavioral child neurology, funded with a Researcher-in-Training Award from the Child Neurology Foundation , working with Harvard developmental neuroscientist Charles Nelson . In Nelson's lab, she learned methods for assessing brain activity -- electroencephalography (EEG) and eye tracking -- to help her understand the formation of neural connections in the brains of infants and toddlers.
Earlier this year, Jeste moved into a tenure-track position at the University of California, Los Angeles (UCLA), with a joint appointment in neurology in the medical school and at the Center for Autism Research and Treatment  (CART). At UCLA, Jeste will be performing neurology exams on young patients, elucidating differences in motor skills among patients with autism spectrum disorders. She is also setting up an EEG lab to study cognition in infants and toddlers and adding electrophysiology to the suite of techniques available to UCLA scientists in a multi-investigator study comparing autism patients with their younger siblings.
Jeste's work is just one thread of a complex fabric of research and researchers -- geneticists, developmental psychologists, neuroscientists, clinical psychologists, and physicians -- who have come together to understand autism spectrum disorders, from molecular origins to clinical treatment.
Autism is a developmental disability that affects social skills, motor skills, and language. Symptoms, which typically start before the age of 3, include repetitive behaviors and problems communicating and interacting with others. Autism's complexity -- and the need for a robust collaborative approach -- was part of the motivation behind the National Institutes of Health's (NIH's) collaborative funding strategy that started in the late 1990s with the creation of Collaborative Programs of Excellence in Autism and then the Studies to Advance Autism Research and Treatment Network in the early 2000s. Those programs merged in 2006 to create the Autism Centers of Excellence (ACE) program , which funds diverse teams of researchers and their interdisciplinary projects, either at single institutions or in affiliated networks.
UCLA's CART is one of those NIH-funded centers. UCLA's work on autism dates back to the 1950s, but the earlier NIH funding mechanisms allowed UCLA to organize those programs into a single, cross-disciplinary translational research program in 2003, says Susan Bookheimer , the principal investigator on CART's current ACE center grant.
Bookheimer is a clinical psychologist who uses functional magnetic resonance imaging (fMRI) to explore the organization of language and memory in the brain. Her clinical experience and imaging expertise serve as a translational bridge on many of CART's projects. She often does brain imaging on autism patients before and after particular treatments. She works with geneticist Daniel Geschwind , CART's director, to compare genetic data from individuals with known autism risk factors with fMRI images of their brains. In addition, she and her colleague Mirella Dapretto  are doing brain scans to monitor the effects of a behavioral-intervention study carried out by UCLA's Connie Kasari .
The development of genetic tests and brain imaging during the past 30 years -- and a more interdisciplinary approach -- has boosted autism research, says Sally Rogers , a developmental psychologist at UC Davis's Medical Investigation of Neurodevelopmental Disorders  (MIND) Institute. "As the new autism research groups emerge, they're much more interdisciplinary."
One part of Rogers's work is a large study, with David Amaral , a neuroscience colleague, called the Autism Phenome Project , which combines brain imaging, genomics, behavior, physiology, and epidemiology data from hundreds of children with autism spectrum disorders and compares that information with data from children who've developed typically. By looking at the same group of children through these different scientific lenses, the collaborators hope to better understand the roots of autism, categorize different autism subtypes, and devise preventions and treatments. "It takes a lot of money, a lot of families, a lot of scientists, and a lot of time" to do get this type of information, Rogers says.
The NIH budget for autism research has risen steadily as awareness of the disease -- and its apparent prevalence -- has increased. NIH's autism-research budget was $132 million in 2009 and got a $64 million boost from Recovery Act funding. Bigger budgets have allowed interdisciplinary, multi-investigator autism research to flourish and seeded the creation of interdisciplinary, collaborative training programs, including the one Rogers and Amaral oversee at UC Davis's MIND Institute.
The increasingly interdisciplinary nature of autism research means that it's possible to attack the disorder from many perspectives. Yet experts agree that time in the clinic -- even for basic science -- is an important part of the training experience. "Try to interact with the patients and really see what these families and these patients go through," advises David Shirinyan, a neuroscience postdoc at UCLA who has worked part-time in an autism clinic for the past 8 years. "It's really one thing to read about a set of symptoms and symptom clusters, but it's a whole other issue when you see it actually play out in somebody's life." Balancing his 10- to 15-hour per week clinic commitment with his research responsibilities has been challenging, he says, but the work in the clinic has helped him develop his research focus: using fMRI to understand the reward circuitry in the brain. That's important, he says, because success in behavioral treatments often depends on finding rewards that motivate patients.
Postdoctoral training can be an excellent time to take a step back from your specialty to learn about autism more broadly and seek a focus for your research program. Neuroscientist Melissa Bauman  gained that perspective through the autism research training program , which Rogers and Amaral run at U.C. Davis's MIND Institute. Although her graduate research on the development of the amygdala had implications for studying autism, Bauman says, "I knew very little about this disorder that I was attempting to model" before she started her postdoctoral fellowship. The fellowship included 2 years of coursework in child development, neuroscience, immunology, genetics, and general research design, among other fields. Bauman also spent time observing a clinician, Sally Ozonoff , who worked with children.
Jeste already had clinical experience, so her priority was to find funding and mentoring that would allow her to devote time to research. During her year as a chief resident, she and Nelson wrote a book chapter on autism, which allowed her to immerse herself in the autism literature. She wrote her Child Neurology Foundation proposal that year, which set her up for her next training phase. If you wait until you complete your clinical training to apply for research funding, Jeste warns, "sometimes you're stuck, and you don't have the funding for what you want to do."
Despite the increase in funding, researchers who enter the field still face a daunting task: going on the academic job market. Tenure track positions are scarce, particularly at the large interdisciplinary centers. So some early-career researchers are staying on at these large centers -- but outside the tenure track and using soft money to build their research programs.
Bauman took this path, staying on in a non-tenure track research appointment at UC Davis's medical school after completing her postdoctoral training. Her research -- using mouse and nonhuman primate models to examine how a mother's immune environment might contribute to the risk of a child with autism -- relies on the animal-care facilities and other infrastructure available at UC Davis. "That limits my willingness to relocate for other tenure track options," she says.
Large centers do have advantages, notes Michael Siller , who completed his Ph.D. and postdoctoral work at UCLA and then went a different route. In 2007, he took a tenure-track position at the much smaller Hunter College , part of the City University of New York. "I felt like I wanted to be able to stand on my own feet," he says. But a smaller institution, he has learned, means more limited resources, which has slowed his research pace.
If you want to take an independent path -- particularly if you end up at a smaller institution -- you need to be strategic in learning all the skills you'll need to run your lab, Siller advises. For his behavioral treatment studies, which focus on developing communication skills through play-interactions between a parent and a child, he's had to learn how to make and confirm an autism diagnosis. When he was at UCLA, other researchers handled diagnosis, and he could focus on the intervention work. In his own lab, Siller has to understand how to do both.
Although Siller has an ongoing collaboration with one of his UCLA colleagues, finding new collaborators has been a challenge, he says. And without built-in connections to agencies that serve autistic children and their families and a medical school affiliation, Siller has had a hard time recruiting subjects for his studies. Recruitment for a recent study of his took more than a year and hundreds of phone calls. "The funding is so tied to these centers that it's hard to get funding outside of that structure," he says. "Everything you do is so dependent on large samples, and you can't get large samples in small places."
Researchers in this field are driven not only by their intellectual curiosity but also by their desire to help patients and their families. "Everyone knows someone who has a child with autism, and that's very motivating for many scientists," Bauman says. "You see parents struggling, and it really hits home."
Early experience working with a child in a residential facility for children with developmental disorders in his native Germany set Siller on his research path. "It made me very curious, and it felt like very few people have many answers," he says. With the challenges of the research path, that passion is important. The work can be tedious, and the payoff from a study might be 10 years down the road. Yet, "I love this field. I wouldn't want to do anything else."
The centers funded through NIH's Autism Centers of Excellence program offer a place to start to see what kind of research is going on in autism:
- Center for Autism Research and Treatment  at the University of California, Los Angeles
- Autism Center of Excellence  and Healthy Infant Development Laboratory at the University of California, San Diego
- Autism Center of Excellence  at the University of Illinois, Chicago
- Autism Center  at the University of Washington
- Autism Center of Excellence  at the University of Pittsburgh
- The Yale Autism Program 
Photo (top): Mykl Roventine 
Sarah A. Webb  writes from Brooklyn, New York.