announcement

Read about Simons Searchlight’s commitment to data privacy and security.

Publications

Date Revised: December 2025

Thank you to all the families for participating in Simons Searchlight. Through your involvement, we aim to assist researchers and geneticists worldwide in understanding genetic disorders affecting you or your family.

The research conducted using Simons Searchlight data has resulted in numerous published papers. These papers undergo a peer-review process, where other scientists assess and validate the research before publication in scientific journals. Additionally, some findings are shared via preprints, allowing rapid dissemination of information to the scientific community.

Many of the publications feature the name “Simons Variation in Individuals Project” (SimonsVIP), which was the original name of our research program, now known as Simons Searchlight.

The listed articles are organized from newest to oldest. You can explore publications by specific genetic conditions using the categories below.

As of December 2025, Simons Searchlight has contributed to 135 publications and preprints, and we will continue to summarize new publications.

For accessibility, the Simons Foundation encourages researchers to make their publications open access. If you cannot access a journal article, we recommend reaching out to the last author listed on the paper to request a copy.

Understanding Publication Reference Titles:

-The article title is followed by publication details, including where and when it was published.
– If there are more than three authors, we use “et al.” to represent additional contributors.
– Journals are referenced using shorthand names.

Disclaimer: Please be aware that papers posted on medRxiv (pronounced med-archive) or bioRxiv (pronounced bio-archive) are not peer-reviewed or edited before online publication. In contrast, all other articles listed here have undergone review by fellow researchers to ensure quality and accuracy. While posting on medRxiv or bioRxiv allows researchers to share findings quickly, the final published results may differ after undergoing formal peer review for journal publication.

Show More
Show Less

Search

  • Filter
  • Clear All
Genetic Condition
Year of Publication
134 Publications
Challenges in multi-task learning for fMRI-based diagnosis: Benefits for psychiatric conditions and CNVs would likely require thousands of patients
  • The researchers aimed to use computer machine learning to see if it is possible to detect and diagnose certain genetic conditions from brain imaging. They collected resting-state functional magnetic resonance imaging (rs-fMRI) data for 7 different copy number variants. Show More
  • Imaging data from Simons Searchlight participants with a 16p11.2 deletion, 16p11.2 duplication, 1q21.1 deletion, and 1q21.1 duplication were included in this research, as well as data from other research studies.
  • rs-fMRI images from 2,872 individuals were included from people with a 1q21.1 deletion, 1q21.1 duplication, 15q11.2 deletion, 16p11.2 deletion, 16p11.2 duplication, 22q11.2 deletion, and 22q11.2 duplication. These images were compared to images from people with no genetic diagnosis. The researchers also obtained images from people diagnosed with attention-deficit/hyperactivity disorder (ADHD), autism, schizophrenia, and bipolar disorder.
  • Once the researchers completed the machine learning process, they compared their results with images from the UK Biobank, which contains data from 30,185 people.
  • The researchers described the method used, called multi-task learning, and how they applied it to the various image datasets.
  • The researchers found that overall, multi-task learning might be a possible predictor of the sex and age of a person with no genetic diagnosis in the UK Biobank.
  • The researchers found that it was difficult to use machine learning to diagnose ADHD, autism, schizophrenia, bipolar disorder, or any of the copy number variants tested in this study. 22q11.2 deletion had the highest accuracy (about 90 percent), whereas schizophrenia, bipolar disorder, 16p11.2 deletion, 16p11.2 duplication, and 1q21.1 deletion reached over 70 percent accuracy, in one particular model tested. The findings were not always consistent in the other types of machine learning models.
  • The researchers indicated that the sample size of the image set used for machine learning was a very important factor for detecting a diagnosis correctly. Show Less
Imaging Neurosci 2, 1-20 (2024)
Harvey et al.
Full Article

16p11.2 deletion
16p11.2 duplication
1q21.1 deletion
1q21.1 duplication
2024

A comparison of symptom profiles in probands with 16p11.2 deletion and duplication syndromes: Repetitive behavior and psychosis proneness
  • Simons Searchlight data was used at Bucknell University in an honors thesis. Show More
  • The student looked at repetitive behavior and psychosis spectrum behavior in 116 people with a 16p11.2 deletion and 59 people with a 16p11.2 duplication. The study also included 32 people with no genetic diagnosis.
  • The student used information from 3 surveys: the Childhood Routines Inventory-Revised, the Childhood Oxford-Liverpool Inventory of Feelings and Experiences, and the Child Behavior Checklist.
  • The student found that, in general, people with a 16p11.2 duplication had more severe maladaptive repetitive behaviors and psychosis symptoms than people with a 16p11.2 deletion.
  • People with a 16p11.2 duplication more often had repetitive, compulsive-like behaviors and were prone to psychosis.
  • People with a 16p11.2 deletion had more issues with socializing, which was suggested to be a sign of autism in this group.
  • Finally, males with a 16p11.2 duplication were more affected on these surveys compared with females who had a 16p11.2 duplication. No differences were found between males and females with a 16p11.2 deletion.
  • Simons Searchlight is excited that these data are being used by young researchers, perhaps inspiring them to do more research on neurodevelopmental conditions. Show Less
Bucknell University, Honors Theses 686, (2024)
Rakauskas
Full Article

16p11.2 deletion
16p11.2 duplication
2024

Motor phenotypes associated with genetic neurodevelopmental disorders
  • The researchers studied the motor abilities of Simons Searchlight participants with genetic neurodevelopmental disorders. Show More
  • This study included 959 Simons Searchlight participants with 57 unique genetic conditions. The average ages for developmental milestones were described for people with autism and no intellectual disability (ID), people with autism and ID, and people with ID without autism. For genetic communities where the clinical spectrum was outside the Simons Searchlight community average, the researchers provide more details about that community.
  • The average age that people learned to sit unsupported was 11 months, and people with ID without autism learned to sit later on average (13 months). People with autism and no ID or autism and ID learned to sit around 10 months.
  • People with GRIN1-related syndrome learned to sit at an average age of 26 months.
  • The average age of walking in the Simons Searchlight study was 25 months. Children with ID without autism had a later average age of walking, about 29 months. Children with autism and no ID were walking around 23 months of age, and children with autism and ID were walking by the age of 26 months.
  • The average age of walking for people with CTNNB1-related syndrome was around 37 months. For people with HNRNPH2-related syndrome, it was around 42 months.
  • Some people with the following genetic conditions were not yet walking over the age of 1.5 years: ARID1B, CHAMP1, CTNNB1, GRIN1, GRIN2B, HNRNPH2, PPP2R1A, PPP2R5D, SCN2A, STXBP1, and TBR1.
  • The majority of people in Simons Searchlight reported having low muscle tone – 577 out of 696 people or 83 percent.
  • About 10 percent of people (73 out of 696 people) reported being diagnosed with cerebral palsy (CP). The following genetic conditions had at least one person with CP: ADNP, ASXL3, AUTS2, CHAMP1, CSNK2A1, CTNNB1, DDX3X, DYRK1A, FOXP1, GRIN2B, HIVEP2, HNRNPH2, IRF2BPL, PPP2R1A, PPP2R5D, PTCHD1, SCN2A, SETBP1, SETD5, STXBP1, and SYNGAP1.
  • The researchers explained that CP is a descriptive term and does not explain why a person has a CP diagnosis. People should still have genetic testing when they have CP, and one does not contradict the other. Show Less
Ann Clin Transl Neurol Epub ahead of print, (2024)
Almansa et al.
Full Article

All Genes
2024

Etiology and molecular mechanisms of PPP2R5D-related developmental disorders
  • This thesis research used a PPP2R5D induced pluripotent stem cell (iPSC) line from the Simons Foundation biorepository. Show More
  • iPSCs are a special type of cells that can be turned into other body cells, making it easier to do research on parts of the body that are difficult to study, such as brain cells.
  • The student investigated the effects of a p.Glu198Lys in one cell line and the effects of p.Glu420Lys in a different cell line.
  • The student found that cell lines with these variants had different sizes, shapes, and rates of growth compared with cell lines without neurodevelopmental variants.
  • Their research highlights that PPP2R5D plays multiple roles in the cell. Show Less
University of South Alabama PhD Thesis, 206 (2024)
Li
Full Article

PPP2R5D
2024

Analysis of mitochondrial DNA replisome in autism spectrum disorder: Exploring the role of replisome genes
  • The researchers investigated how energy production in the cells might be different in autistic people. In cells, energy is transferred from food molecules to the molecules that the cell can use to do work. This process happens inside a structure called the mitochondria. Show More
  • The researchers used a Simons Searchlight 16p11.2 deletion induced pluripotent stem cell (iPSC) line from a person who had an autism diagnosis.
  • From cheek swab samples of children with autism, the researchers found that children with autism had a different number of mitochondrial DNA. This finding suggests that the production of mitochondria is different in children with autism.
  • The researchers tested for oxidative stress, which is a byproduct of turning food into energy. Children with autism had a higher amount of oxidative stress and an inflammation marker in their cheek samples. The researchers suggested that oxidative stress might contribute to mitochondria dysfunction and inflammation might play a role in immune dysregulation in autistic people.
  • The researchers also found that the iPSCs of a person with autism were more prone to mitochondrial errors under a stressed condition (induced by a drug) compared with cells from a person without autism.
  • The researchers did not find links between mitochondrial gene regulation and brain volume, other genetic variants in certain mitochondrial genes, or the development of autistic features. Show Less
Autism Res Epub ahead of print, (2024)
Rojas et al.
Full Article

16p11.2 deletion
2024

Rare CNVs and phenome-wide profiling highlight brain structural divergence and phenotypical convergence
  • These researchers compared the brain structures of people with copy number variants (CNVs) and people in the UK Biobank that were from the general population. This study included eight CNVs: deletions or duplications of 1q21.1, 15p11.2, 16p11.2, and 22q11.2. Show More
  • This study included participants from several research studies or universities: Simons Searchlight; Cardiff University; 16p11.2 European Consortium; University of Montreal; and University of California, Los Angeles. There were 548 people with a CNV and 312 people with no genetic condition.
  • The researchers used computer analytics to study the brain features of each of the CNVs and created one of the largest brain imaging studies to date.
  • They made a comparison between brain structures and medical features in order to find links between the two. They aimed to understand how brain structure can lead to behaviors.
  • Brain volumes were smaller in participants with a 1q21.1 deletion, 15p11.2 duplication, 16p11.2 duplication, and 22q11.2 deletion. Brain volumes were bigger in people with a 1q21.1 duplication, 15p11.2 deletion, 16p11.2 deletion, and 22q11.2 duplication.
  • They found that people with a 16p11.2 deletion or 22q11.2 deletion had unique brain patterns, whereas people with a 15p11.2 duplication had brain structures that were similar to the general population.
  • People with a 16p11.2 deletion had the highest number of brain regions affected, whereas people with a 15p11.2 duplication had the lowest number of regions affected.
  • The researchers created a graph showing the large-scale network of each CNV, and they studied characteristics of people with a CNV, such as body size, lifestyle, and blood factors. All eight CNVs had strong associations with diastolic blood pressure, a protein called alkaline phosphatase, and red blood cell count.
  • This research was supported by a grant from the Simons Foundation Autism Research Initiative (SFARI). Show Less
Nat Hum Behav Epub ahead of print, (2023)
Kopal et al.
Full Article

16p11.2 deletion
16p11.2 duplication
1q21 deletion
1q21 duplication
2023

Subcortical brain alterations in carriers of genomic copy number variants
  • Researchers know that copy number variants (CNV) can contribute to neurodevelopmental and psychiatric disorders, such as autism and schizophrenia. A CNV happens when there is a change in a section of DNA that results in a gene or several genes being deleted or duplicated. 16p11.2 deletion is an example of a CNV. Show More
  • The goal of this study was to compare the sizes and patterns of regions of the brain for 11 different CNVs.
  • The researchers studied the following deletions and duplications: 1q21.1, 13q12.12, 15q11.2, 16p11.2, 16p13.11, and 22q11.2. They also studied duplications within chromosome 1, including the TAR region. This study included 675 participants. Brain MRI images for 1q21.1 and 16p11.2 CNVs came from Simons Searchlight participants.
  • The researchers found that CNVs with more genes deleted or duplicated within that region, such as 16p11.2, resulted in more structural changes on magnetic resonance imaging (MRI). All 11 CNVs had changes in the thickness of the brain region that affects cognitive, affective, and social functions. The researchers did not find a link between overall brain volume and the cognition of the person with these CNVs, but, the shape/structure of the brain regions were linked to a person’s cognition.
  • People with 16p11.2 duplications had brain structure changes that were different than those seen in people with autism, and people with no genetic diagnosis. This suggests that there may be several different brain changes that could lead to autism.
  • This research was supported by a grant from the Simons Foundation Autism Research Initiative (SFARI). Show Less
Am J Psychiatry 180, 685-698 (2023)
Kumar et al.
Full Article

16p11.2 deletion
16p11.2 duplication
16p13.11 deletion
1q21.1 deletion
1q21.1 duplication
2023