Biology subjects

Bhandari, A.

Publications and source records attributed to Bhandari, A..

25 records · Page 2Linked to original sources

bioRxiv2020

Noise correlations for faster and more robust learning.

Distributed population codes are ubiquitous in the brain and pose a challenge to downstream neurons that must learn an appropriate readout. Here we explore the possibility that this learning problem is simplified through inductive biases implemented by stimulus-independent noise correlations that constrain learning to task-relevant dimensions. We test this idea in a set of neural networks that learn to perform a perceptual discrimination task. Correlations among similarly tuned units were manipulated independently of overall population signal-to-noise ratio in order to test how the format of stored information affects learning. Higher noise correlations among similarly tuned units led to faster and more robust learning, favoring homogenous weights assigned to neurons within a functionally similar pool, and could emerge through Hebbian learning. When multiple discriminations were learned simultaneously, noise correlations across relevant feature dimensions sped learning whereas those across irrelevant feature dimensions slowed it. Our results complement existing theory on noise correlations by demonstrating that when such correlations are produced without significant degradation of the signal-to-noise ratio, they can improve the speed of readout learning by constraining it to appropriate dimensions. Significance statementPositive noise correlations between similarly tuned neurons theoretically reduce the representational capacity of the brain, yet they are commonly observed, emerge dynamically in complex tasks, and persist even in well-trained animals. Here we show that such correlations, when embedded in a neural population with a fixed signal to noise ratio, can improve the speed and robustness with which an appropriate readout is learned. In a simple discrimination task such correlations can emerge naturally through Hebbian learning. In more complex tasks that require multiple discriminations, correlations between neurons that similarly encode the task-relevant feature improve learning by constraining it to the appropriate task dimension.

neuroscience↗

bioRxiv2020

Fronto-parietal, cingulo-opercular and striatal contributions to learning and implementing control policies

Efficient task performance requires co-ordination of internal cognitive processes by implementing control policies adapted to the dynamic structure of task demands. The cognitive and neural basis of control policy implementation remains poorly characterized, in part because it is typically confounded with implementing new stimulus-response rules. To disambiguate these processes, we asked participants to perform multiple novel variants of a working memory control task. Each variant had a unique, novel sequential trial structure, but all shared common stimulusresponse rules, enabling us to test control policy implementation separate from rule learning. Behaviorally, we found evidence for two adaptive processes tied to control policy implementation. One process was reflected in slower responses on the first trial with a novel sequential trial structure, followed by rapid speeding on subsequent trials. A second process was reflected in the diminishing size of the first trial cost as participants accommodated different variants of the task over many blocks. Using fMRI, we observed that the striatum and a cingulo-opercular cortical network increased activity to the first trial, tracking the fast adjustment. This pattern of activity dissociated these regions from a fronto-parietal network including dorsolateral PFC, inferior frontal junction, inferior parietal sulcus, and rostrolateral PFC, which showed a slower decline in activity across trials, mirroring findings in rule implementation studies, but in the absence of rule implementation demands. Our results reveal two adaptive processes underlying the implementation of efficient, generalizable control policies, and suggest a broader account of the role of a cortico-striatal network in control policy implementation. Significance statementRapid adaptation to novel tasks is a hallmark of human behavior. Understanding how human brains achieve this is of critical importance in neuroscience. Here we broaden the scope of this problem, going beyond task rules to more broadly consider the cognitive control demands produced by novel task dynamics. We propose that humans rely on two adaptive processes to rapidly implement efficient, generalizable control policies as task dynamics change, even when task rules remain unchanged. One process unfolds rapidly and underlies efficient adaptation. A second process unfolds slowly with experience across task conditions and underlies generalization of control policies. Using fMRI, we identify cingulo-opercular cortex, fronto-parietal cortex and striatum as dissociable components of a cortico-striatal network that contribute to control implementation.

neuroscience↗

bioRxiv2019

Early-Stage Ocular Hypertension Alters Retinal Ganglion Cell Synaptic Transmission in the Visual Thalamus

Axonopathy is a hallmark of many neurodegenerative diseases including glaucoma, where elevated intraocular pressure (ocular hypertension, OHT) stresses retinal ganglion cell (RGC) axons as they exit the eye and form the optic nerve. OHT causes early changes in the optic nerve such as axon atrophy, transport inhibition, and gliosis. Importantly, many of these changes appear to occur prior to irreversible neuronal loss, making them promising points for early diagnosis of glaucoma. It is unknown whether OHT has similarly early effects on the function of RGC output to the brain. To test this possibility, we elevated eye pressure in mice by anterior chamber injection of polystyrene microbeads. 5 weeks post-injection, bead-injected eyes showed a modest RGC loss in the peripheral retina, as evidenced by RBPMS antibody staining. Additionally, we observed reduced dendritic complexity and lower spontaneous spike rate of On-RGCs, targeted for patch clamp recording and dye filling using a Opn4-cre reporter mouse line. To determine the influence of OHT on retinal projections to the brain, we expressed Channelrhodopsin-2 (ChR2) in melanopsin-expressing retinal ganglion cells by crossing the Opn4-cre mouse line with a ChR2-reporter mouse line and recorded post-synaptic responses in thalamocortical relay neurons in the dorsal lateral geniculate nucleus (dLGN) of the thalamus evoked by stimulation with 460 nm light. The use of a Opn4-cre reporter system allowed for expression of ChR2 in a narrow subset of RGCs responsible for image-forming vision in mice. Five weeks following OHT induction, paired pulse and high-frequency stimulus train experiments revealed that presynaptic vesicle release probability at retinogeniculate synapses was elevated. Additionally, miniature synaptic current frequency was slightly reduced in brain slices from OHT mice and proximal dendrites of post-synaptic dLGN relay neurons, assessed using a Sholl analysis, showed a reduced complexity. Strikingly, these changes occurred prior to major loss of RGCs labeled with the Opn4-Cre mouse, as indicated by immunofluorescence staining of ChR2-expressing retinal neurons. Thus, OHT leads to pre- and post-synaptic functional and structural changes at retinogeniculate synapses. Along with RGC dendritic remodeling and optic nerve transport changes, these retinogeniculate synaptic changes are among the earliest signs of glaucoma.

neuroscience↗

bioRxiv2017

Phenome-wide association studies (PheWAS) across large "real-world data" population cohorts support drug target validation

Phenome-wide association studies (PheWAS), which assess whether a genetic variant is associated with multiple phenotypes across a phenotypic spectrum, have been proposed as a possible aid to drug development through elucidating mechanisms of action, identifying alternative indications, or predicting adverse drug events (ADEs). Here, we evaluate whether PheWAS can inform target validation during drug development. We selected 25 single nucleotide polymorphisms (SNPs) linked through genome-wide association studies (GWAS) to 19 candidate drug targets for common disease therapeutic indications. We independently interrogated these SNPs through PheWAS in four large \"real-world data\" cohorts (23andMe, UK Biobank, FINRISK, CHOP) for association with a total of 1,892 binary endpoints. We then conducted meta-analyses for 145 harmonized disease endpoints in up to 697,815 individuals and joined results with summary statistics from 57 published GWAS. Our analyses replicate 70% of known GWAS associations and identify 10 novel associations with study-wide significance after multiple test correction (P<1.8x10-6; out of 72 novel associations with FDR<0.1). By leveraging directionality and point estimate of the effect sizes, we describe new associations that may predict ADEs, e.g., acne, high cholesterol, gout and gallstones for rs738409 (p.I148M) in PNPLA3; or asthma for rs1990760 (p.T946A) in IFIH1. We further propose how quantitative estimates of genetic safety/efficacy profiles can be used to help prioritize candidate targets for a specific indication. Our results demonstrate PheWAS as a powerful addition to the toolkit for drug discovery.\n\nOne Sentence SummaryMatching genetics with phenotypes in 800,000 individuals predicts efficacy and on-target safety of future drugs.

genetics↗

bioRxiv2017

A model species for agricultural pest genomics: the genome of the Colorado potato beetle, Leptinotarsa decemlineata (Coleoptera: Chrysomelidae)

The Colorado potato beetle is one of the most challenging agricultural pests to manage. It has shown a spectacular ability to adapt to a variety of solanaceaeous plants and variable climates during its global invasion, and, notably, to rapidly evolve insecticide resistance. To examine evidence of rapid evolutionary change, and to understand the genetic basis of herbivory and insecticide resistance, we tested for structural and functional genomic changes relative to other arthropod species using genome sequencing, transcriptomics, and community annotation. Two factors that might facilitate rapid evolutionary change include transposable elements, which comprise at least 17% of the genome and are rapidly evolving compared to other Coleoptera, and high levels of nucleotide diversity in rapidly growing pest populations. Adaptations to plant feeding are evident in gene expansions and differential expression of digestive enzymes in gut tissues, as well as expansions of gustatory receptors for bitter tasting. Surprisingly, the suite of genes involved in insecticide resistance is similar to other beetles. Finally, duplications in the RNAi pathway might explain why Leptinotarsa decemlineata has high sensitivity to dsRNA. The L. decemlineata genome provides opportunities to investigate a broad range of phenotypes and to develop sustainable methods to control this widely successful pest.

genomics↗

bioRxiv2017

Learning And Transfer Of Working Memory Gating Policies

Abstract knowledge about the tasks we encounter enables us to rapidly and flexibly adapt to novel task contexts. Previous research has focused primarily on abstract rules that leverage shared structure in stimulus-response (S-R) mappings as the basis of such task knowledge. Here we provide evidence that working memory (WM) gating policies - a type of control policy required for internal control of WM during a task - constitute a form of abstract task knowledge that can be transferred across contexts. In two experiments, we report specific evidence for the transfer of selective WM gating policies across changes of task context. We show that this transfer is not tied to shared structure in S-R mappings, but instead in the dynamic structure of the task. Collectively, our results highlight the importance of WM gating policies in particular, and control policies in general, as a key component of the task knowledge that supports flexible behavior and task generalization.

neuroscience↗

bioRxiv2017

Just above chance: is it harder to decode information from human prefrontal cortex BOLD signals?

Understanding the nature and form of prefrontal cortex representations that support flexible behavior is an important open problem in cognitive neuroscience. In humans, multi-voxel pattern analysis (MVPA) of fMRI BOLD measurements has emerged as an important approach for studying neural representations. An implicit, untested assumption underlying many PFC MVPA studies is that the base rate of decoding information from PFC BOLD activity patterns is similar to that of other brain regions. Here we estimate these base rates from a meta-analysis of published MVPA studies and show that the PFC has a significantly lower base rate for decoding than visual sensory cortex. Our results have implications for the design and interpretation of MVPA studies of prefrontal cortex, and raise important questions about its functional organization.

neuroscience↗