About

Gustavo D. Aguirre is an Emeritus Professor of Clinical Studies - Philadelphia and a Professor of Medical Genetics and Ophthalmology at the University of Pennsylvania School of Veterinary Medicine. His research is directed towards identifying the genetic causes of inherited blindness, understanding the mechanisms linking mutations to disease, and developing treatment approaches. His studies are conducted in the dog model, which exhibits a large variety of inherited photoreceptor diseases that are being characterized at the molecular level. His work involves disease hunting through phenotype-directed candidate gene analysis and genome-wide scans to identify disease-bearing chromosomal regions, followed by positional cloning and mutation identification. Once the causative genes and mutations are identified, he employs molecular and proteomic approaches to examine disease mechanisms. Concurrently, he develops gene-based therapies aimed at restoring function and preventing degeneration of mutant photoreceptor cells. Key projects include investigating the mechanisms of disease and treatment for X-linked retinitis pigmentosa caused by RPGR mutations, studying the effects of light damage on rhodopsin mutations, and characterizing Bestrophin mutations related to Best Macular Dystrophy. His research also explores the distribution and function of proteins involved in retinal diseases, with the goal of advancing therapeutic strategies for inherited retinal degenerations.

Research topics

• Computer Science
• Artificial Intelligence
• Data Mining
• Medicine
• Genetics
• Computer vision
• Database
• Neuroscience
• Biology
• Pathology
• Cell biology
• Molecular biology
• Computational biology
• Physics
• Bioinformatics
• Radiology
• Chemistry

Selected publications

• Temporal dynamics of human color processing measured using a continuous tracking task

  Journal of Vision · 2025-02-27 · 2 citations

  articleOpen access

  We characterized the temporal dynamics of color processing using a continuous tracking paradigm by estimating subjects' temporal lag in tracking chromatic Gabor targets. To estimate the lag, we computed the cross-correlation between the velocities of the Gabor target's random walk and the velocities of the subject's tracking. Lag was taken as the time of the peak of the resulting cross-correlogram. We measured how the lag changes as a function of chromatic direction and contrast for stimuli in the LS cone contrast plane. In the same set of subjects, we also measured detection thresholds for stimuli with matched spatial, temporal, and chromatic properties. We created a model of tracking and detection performance to test whether a common representation of chromatic contrast accounts for both measures. The model summarizes the effect of chromatic contrast over different chromatic directions through elliptical isoperformance contours, the shapes of which are contrast independent. The fitted elliptical isoperformance contours have essentially the same orientation in the detection and tracking tasks. For the tracking task, however, there is a striking reduction in relative sensitivity to signals originating in the S cones.

  PublisherOA PDFDOI

• Daily light exposure habits of youth with migraine: a prospective exploratory study

  npj Biological Timing and Sleep · 2025-11-25 · 2 citations

  articleOpen accessSenior author

  Photophobia is a common symptom in youth with migraine, but it is unknown if this leads to light-avoidant behavior, and its effect on migraine disease burden. We conducted an exploratory study measuring light exposure using wearable light logger pendants in 20 adolescents with migraine (ages 15–21) over one week. On average, participants received recommended light exposure during only 15.1% during the day ( $$\ge$$ 250 lux melanopic equivalent daytime illuminance from 7:00–17:00) but were generally within recommended light levels in the evening (78.1% from 20:00–23:00 at $$\le$$ 10 lux), and at night (99.1% from 0:00–6:00 at $$\le$$ 1 lux). We observed differences in light exposure on weekdays versus weekends. Individual variation in circadian delay of light exposure was strongly correlated with the number of headache days per month (Rho = 0.66, p = 0.002). Measuring daily light exposure is feasible in pediatric populations with photophobia and warrants further study.

  PublisherOA PDFDOI

• Evolution of Visual Field Defects After Occipital Stroke: A Quantitative Analysis

  Translational Vision Science & Technology · 2025-06-06 · 1 citations

  articleOpen access

  Purpose: To quantitatively re-evaluate the natural progression of homonymous visual defects in occipital stroke patients using metrics derived from automated perimetry. Methods: We performed a retrospective chart review of stroke patients with homonymous hemianopia who underwent repeated, monocular Humphrey perimetry using the 24-2 SITA standard pattern from 2011 to 2019 at a large US academic medical center. Reliable tests in either eye were analyzed to track progression of the perimetric mean deviation (PMD) and deficit area before and after a temporal boundary set at 6 ± 1 months post-stroke. Results: Out of 532 patients diagnosed with homonymous hemianopia, repeated Humphrey tests meeting reliability criteria were available for 73 individuals, who ranged from 0 to 2986 days post-stroke. Over 77% of these patients experienced spontaneous perimetric improvements in the subacute post-stroke period (<6 months), much larger for partial than complete visual field defects. In the chronic period (>6 months), perimetric measurements were stable overall, but spontaneous changes > ±1 dB in PMD were seen in about half of the cohort. Conclusions: Quantitative analysis of Humphrey perimetry confirmed prior reports of prevalent spontaneous improvements early after occipital stroke in adults. Though much less frequent, improvements as well as worsening of the visual defects still occurred in the chronic period. Translational Relevance: Our quantification of perimetric changes over time after occipital stroke provides an empirical and methodological basis for future studies aiming to better understand and prognostically manage patient expectations regarding recovery. It also provides a necessary benchmark against which to contrast the impact of restorative therapies being developed for this condition.

  PublisherDOI

• Encoding fidelity of flicker frequency does not differ by post-receptoral direction or photopic light level

  Journal of Vision · 2025-07-15

  articleOpen accessSenior author

  Temporal sensitivity functions recorded from ganglion cells and post-retinal sites differ for achromatic and chromatic modulations, and at high and low light levels. Visual experience statistics, however, suggest that the 1/f^2 distribution of wide-field temporal variation has the same form for achromatic and chromatic modulations (JH van Hateren et al 2002 J Neurosci; data not available across light levels). Efficient coding models therefore predict similar encoding fidelity across frequency for achromatic and chromatic modulations, despite differences in neural tuning. We tested this idea in a delayed flicker frequency estimation task, using luminance and L–M modulations against high and low photopic backgrounds. METHODS: On each of 1200 trials five participants were shown a 2s “reference” flicker (30° wide, uniform-field, bimodal sinusoidal temporal flicker, 1–32Hz uniform log random sampling). After a 2s delay, the participant adjusted the frequency of a “test” flicker to match their memory of the reference (starting point randomly selected within ±7dB of the reference). Positive feedback was provided for responses within ±1.5dB of the reference. Blocks of trials within a session alternated between LMS (33% contrast) and L–M (7.5%) modulations, and the six sessions alternated between a high (3400cd/m^2) and low (2cd/m^2) luminance background. RESULTS: Variance in estimation responses was a constant proportion of the reference frequency (except for low luminance L–M >20Hz), and did not differ with light level or post-receptoral direction (1.6dB). Responses were biased across frequency, with a negative slope and an intercept at ~2 Hz (i.e., rapid flicker was reported as slower); this bias did not differ with stimulus condition. These results are consistent with a Bayesian observer model constrained by efficient coding, reflecting a common 1/f^2 temporal frequency prior across our four stimulus conditions. Our results demonstrate a distinction between neural tuning functions and encoding for temporal variation.

  PublisherDOI

• Metabolic stress and early cell death in photoreceptor precursor cells following retinal transplantation

  Stem Cell Research & Therapy · 2025-07-24 · 4 citations

  articleOpen access

  BACKGROUND: Progressive photoreceptor loss in retinal degenerative diseases leads to irreversible vision impairment. Transplantation of human embryonic or induced pluripotent stem cell-derived photoreceptor precursor cells (PRPCs) offers potential for vision restoration. However, substantial early donor cell loss remains a major challenge. This study aims to elucidate the mechanisms underlying early PRPC loss and to evaluate host retinal responses to transplantation. METHODS: PRPCs derived from human embryonic stem cells (hESC)-based retinal organoids were subretinally transplanted into both normal and degenerated canine retinas to investigate the impact of host retinal degeneration on transplant survival and integration. Single-cell RNA sequencing (scRNAseq) was performed on transplanted PRPCs 3 days post-transplantation into normal canine retinas, as well as on host retinal cells to identify molecular pathways associated with early donor cell loss. Non-invasive multimodal retinal imaging and immunohistochemical analyses were conducted to assess PRPC survival, integration, and host immune responses. RESULTS: Despite systemic immunosuppression, extensive early loss of human PRPCs occurred within the first week following xenotransplantation into both normal and degenerated canine retinas, suggesting that factors beyond immune activation contribute to donor cell loss. Transcriptomic analysis identified metabolic stress as a key driver of early donor cell death, characterized by dysregulation of mitochondrial function and oxidative phosphorylation pathways. Microglial infiltration into the donor cell mass was also observed in normal retinas, suggesting a response to donor cell stress and apoptosis. Beyond the initial phase of cell death, surviving donor cells integrated and persisted when transplanted into retinas with a partially preserved outer nuclear layer, whereas cell loss continued when intervention occurred at end-stage degeneration. CONCLUSIONS: Metabolic stress represents a critical barrier to PRPC survival following transplantation. Strategies aimed at enhancing metabolic resilience may improve transplantation outcomes. Furthermore, host retinal responses shape the transplant microenvironment, influencing donor cell survival and integration. These findings highlight the need for targeted interventions to mitigate early metabolic stress and optimize PRPC transplantation strategies for retinal degenerative diseases.

  PublisherOA PDFDOI

• Reconstructing Covert Consciousness

  Neurology · 2025-01-30 · 9 citations

  reviewOpen accessSenior author

  Determining the level of consciousness in patients with brain injury-and more fundamentally, establishing what they can experience-is ethically and clinically impactful. Patient behaviors may unreliably reflect their level of consciousness: a subset of unresponsive patients demonstrate covert consciousness by willfully modulating their brain activity to commands through fMRI or EEG. However, current paradigms for assessing covert consciousness remain fundamentally limited because they are insensitive, rely on imperfect assumptions of functional neuroanatomy, and do not reflect the spectrum of conscious experience. Neural decoding, in which stimuli and concepts are reconstructed from brain activity, offers a novel approach to covert consciousness assessment that overcomes many of these limitations. In this article, we discuss the current state of covert consciousness assessments, their shortcomings, the state of the science in neural decoding, the potential application of neural decoding to disorders of consciousness, and future directions that may help realize this potential. To do so, we searched PubMed and Google Scholar databases for pertinent articles published between January 1990 and September 2024, using the search terms "covert consciousness," "cognitive motor dissociation," "neural decoding," and "semantic decoding." Redefining covert consciousness with neural decoding may improve sensitivity, enhance granularity, and more directly address the question of what patients can experience after brain injury.

  PublisherOA PDFDOI

• Amplifying and ameliorating light avoidance in mice with photoreceptor targeting and calcitonin gene‐related peptide sensitization

  Headache The Journal of Head and Face Pain · 2025-12-15

  articleOpen access

  Abstract Objective The aim of this study was to determine the photoreceptor basis of light avoidance in mice and assess the effect of CGRP sensitization on this behavior. Background Prior studies have suggested that photophobia is mediated by a subset of retinal ganglion cells (RGCs) that contain melanopsin, making them intrinsically photosensitive (ipRGCs). These cells also receive extrinsic input from cones, which can also mediate light sensitivity. Here, we examined how spectral targeting of melanopsin or specific cone types in mice produces light avoidance and whether sensitizing mice with calcitonin gene‐related peptide (CGRP) amplifies the avoidance response to ipRGC stimulation. Methods Light avoidance behavior was measured in a two‐zone chamber illuminated by narrow‐band light‐emitting diodes (LEDs) targeting photopic opsins: 365 nm (ultraviolet [UV]; rodent S‐cone), 460 nm (blue; melanopsin), and 630 nm (red; human L‐cone). In a non‐targeted assay, we assessed the degree of light avoidance in wild‐type (WT) C57BL/6J mice to varying contrasts (0.05 to 1.00) of the blue and red LEDs. In a targeted assay, mice were exposed to zones with differing relative contrast levels (0.50, 0.75, or 1.00) for the targeted photoreceptor(s). This was assessed in transgenic mice with: (1) human L‐cone cone knock‐in (HLCKI) or (2) adult‐onset ablation of M1 ipRGCs (Opn4 aDTA ). Mice were studied without intervention or following chronic intermittent administration of CGRP with either peripheral CGRP or vehicle (Veh) administration every‐other‐day for 9 days. A primary measure (mean +/− SEM) was the asymptote value (AV) of chamber preference. Results WT mice showed greater light avoidance with increasing light contrast. HLCKI mice avoided zones with high melanopsin (1.00: 0.52 ± 0.08; n = 18) and L‐cone (1.00: 0.30 ± 0.11; n = 15) stimulation but showed a preference for the zone with higher S‐cone (1.00: −0.35 ± 0.06; n = 16) stimulation. These effects were contrast‐dependent. The addition of S‐cone stimulation reduced the aversive effect of melanopsin (0.10 ± 0.12; n = 14) or L‐cone (−0.19 ± 0.10; n = 15) contrast. Ablation of ipRGCs in HLCKI x Opn4 aDT A mice eliminated both avoidance of melanopsin stimulation and the preference for S‐cone stimulation, as compared to controls. Nine days of chronic intermittent administration of CGRP led to significantly increased avoidance of melanopsin stimulation (0.58 ± 0.08, n = 21) as compared to Veh administration (0.26 ± 0.09, n = 22) ( F (1, 41) = 5.70, p = 0.022). Conclusions Our findings support a key role for the ipRGCs in the production of photophobia. This aversive response to light stems from integrated ipRGC signals that combine excitatory intrinsic melanopsin and extrinsic L‐cone inputs and are opposed by extrinsic, inhibitory S‐cone input. Chronic elevation of CGRP levels in migraine may amplify ipRGC signals, leading to photophobia.

  PublisherDOI

• Design and calibration of a spectro-temporal light logger

  Journal of Vision · 2025-07-15

  articleOpen accessSenior author

  Measurements of visual experience have largely focused upon spatial visual properties (e.g., orientation, spatial frequency) and have made use of static images and cinematic videos. To support studies of individual differences in visual experience, we created a personalized, wearable, all-day, high-speed light-logger. Our goal is to measure the wide-field, low spatial frequency properties of spectral and temporal visual experience across the full range of photopic vision, and to characterize how these natural statistics interact with eye and head movements. The logger is mounted on inexpensive eye glasses (Pair Eyewear) with custom refractive lenses for each participant; 3D-printed spectacle arms incorporate sensor housings. The World camera (IMX219 chip with an M12 lens) provides 120° FOV, RGB video at 200 FPS (rolling shutter; 8 bit, 30x40 Bayer down-sampled image). The Eye camera (PupilLabs Cam2) provides 120 FPS IR video (global shutter, 400x400). Photopic spectral irradiance is derived from 10 narrow wavelength channels at 1Hz (AS7341 chip); mesopic illuminance is measured for a single wide band (TSL2591 chip). 6-DOF head movement is recorded at 10Hz (LSM6DSV16X chip). A Hall effect sensor detects the presence of magnetic clip-on sunglasses. HDMI and USB cables carry data to a 3D-printed (5.4”x3.4”x2.6”) recording pack worn on a cross-body strap, containing a Raspberry Pi 5, M.2 SSD, and rechargeable battery. We performed sensor calibration against a commercial spectrophotometer (PR670) within a light-integrating sphere receiving output from an 8-channel digital spectral synthesizer (plus varying neutral density filters). Using a custom gain control algorithm, the World camera provides nearly flat temporal sensitivity between 0.25 and 50 Hz across a 4 log illuminance range. Validated spectral sensitivity channels are linear across 5 log illuminance. Eye and World recordings are synchronized to within 10 msecs. We are currently refining the device for subject comfort prior to field trials.

  PublisherDOI

• Network structure influences the strength of learned neural representations

  Nature Communications · 2025-01-24 · 3 citations

  articleOpen access

  From sequences of discrete events, humans build mental models of their world. Referred to as graph learning, the process produces a model encoding the graph of event-to-event transition probabilities. Recent evidence suggests that some networks are easier to learn than others, but the neural underpinnings of this effect remain unknown. Here we use fMRI to show that even over short timescales the network structure of a temporal sequence of stimuli determines the fidelity of event representations as well as the dimensionality of the space in which those representations are encoded: when the graph was modular as opposed to lattice-like, BOLD representations in visual areas better predicted trial identity and displayed higher intrinsic dimensionality. Broadly, our study shows that network context influences the strength of learned neural representations, motivating future work in the design, optimization, and adaptation of network contexts for distinct types of learning. How do humans build mental models from sequences of events? Using fMRI, Kahn et al. show that the network structure of a temporal sequence of stimuli determines the fidelity of event representations and the dimensionality of the representation space.

  PublisherOA PDFDOI

• Identifying chest-worn light logger adherence: a validation study

  medRxiv · 2025-09-30 · 1 citations

  preprintOpen accessSenior author

  Background: Light exposure plays an important role in overall health because it entrains circadian rhythms. Recent technological advances in wearable light loggers allow measurement daily light exposure habits. Using a chest-worn light logger, our goal was to 1) develop methodology for differentiating adherent versus non-adherent use, and 2) define differences in lighting intensity in indoor and outdoor environments, to improve data reliability in future clinical studies using this technology. Methods: Five testers used a 10-channel chest worn light logging device under different conditions of wear and non-wear, and across a variety of indoor and outdoor lighting environments. Measurements from the light logger (photopic illuminance, device orientation, accelerometer data, time of day) were used to train and validate a logistic regression model to differentiate wear from non-wear and correct nighttime placement. This model was then applied to 20 adolescents and young adults with migraine who wore the light logger device for one week. Furthermore, measurements of photopic illuminance and melanopic equivalent daytime illuminance (mEDI) of indoor versus outdoor lighting environments were compared to identify the optimal distinction point between darker indoor and brighter outdoor environments for the chest-worn light logger. Results: Movement, device orientation, light, and time-of-day used as predictors in a logistic regression model had excellent differentiation between wear, non-wear and nighttime use (AUC 0.93 - 0.94), and retained good-to-excellent differentiation when applied to the validation dataset (AUC 0.84 - 0.91). When this model was applied to 20 participants with migraine, we found that 92.9% of participant-days and 71.4% of participant-nights demonstrated at least 80% appropriate use. For differentiating indoor and outdoor lighting environments, the optimal cut-point was 442 lux for photopic illuminance, and 412 lux for mEDI. Conclusions: We demonstrate that internal measurements from a chest-worn light logging device can reliably differentiate wear from non-wear. We also found that the optimal cut-off to differentiate indoor and outdoor lighting environments was similar though slightly lower than then 1,000 lux cut-offs traditionally used to define bright light conditions. These findings can be used to improve data reliability in studies of everyday light exposure in clinical populations using chest-worn light loggers.

  PublisherOA PDFDOI

Recent grants

• NIH Grant U01EY006855

  NIH · $2.9M · 1992

• NIH Grant R01EY013132

  NIH · $3.2M · 2010

• NIH Grant R01EY024681

  NIH · $1.6M · 2021

• NIH Grant K08MH072926

  NIH · $907k · 2009

• NIH Grant R01EY011142

  NIH · $1.3M · 2001

Frequent coauthors

• Gregory M. Acland

  Cornell University

  212 shared

• William A. Beltran

  University of Pennsylvania

  117 shared

• Artur V. Cideciyan

  University of Pennsylvania

  81 shared

• Samuel G. Jacobson

  Penn Presbyterian Medical Center

  79 shared

• David H. Brainard

  University of Pennsylvania

  65 shared

• András M. Komáromy

  Michigan State University

  51 shared

• Simone Iwabe

  Altair Engineering (United States)

  50 shared

• Barbara Zangerl

  Prince Charles Hospital

  49 shared

Labs

• Gustavo D. Aguirre LaboratoryPI

Education

• Other, Veterinary Medicine

  University of Pennsylvania

  1968

• Ph.D.

  University of Pennsylvania

  1975

Awards & honors

• Bestrophin gene mutations cause canine multifocal retinopath…
• Canine RD3 mutation establishes rod cone dysplasia type 2 (r…
• Operating in the dark: a night vision system for surgery in…
• Targeting gene expression to cones with human cone opsin pro…
• Intravitreal injection of ciliary neurotrophic factor (CNTF)…