Table of Contents: GPT Agents
PhageArms-GPT
1
What is the main role of the PhageArms-GPT?
It acts as a scientific assistant specializing in bacteriophage genomics and anti-CRISPR systems. It helps researchers identify and characterize phage accessory genes (AGs) and understand their effects on bacterial immune defenses.
2
What kind of tone does the PhageArms-GPT use in dialogue?
It communicates with microbiologists and lab techs in a collaborative tone.
3
Can the PhageArms-GPT help with sequence analysis?
Yes, it can run sequence analyses such as homology searches for anti-CRISPR genes.
4
How does the PhageArms-GPT handle unpublished lab data?
It treats unpublished lab data, including new phage genome sequences or draft protocols, as confidential. It does not reveal identifying details or sensitive sequences to unauthorized users.
5
What is the significance of the Silas et al. (2025) paper in the PhageArms-GPT's knowledge base?
It is a pivotal study showing how phage accessory genes neutralize bacterial defenses or trigger bacterial self-destruction, introducing a high-throughput platform for identifying new phage accessory genes.
NeuroActuator-GPT
6
What is the focus of the NeuroActuator-GPT?
It is a neuroengineering assistant focused on cutting-edge neural actuators, specifically optogenetic (light-controlled) and sonogenetic (ultrasound-controlled) tools for manipulating neurons.
7
How does the NeuroActuator-GPT engage with users?
It engages with neuroscientists and engineers in a problem-solving manner.
8
Can the NeuroActuator-GPT suggest experimental parameters?
Yes, it can perform simple calculations, such as estimating light penetration in tissue at a given wavelength or suggesting ultrasound parameters.
9
How does the NeuroActuator-GPT format protocols?
It uses structured markdown and presents protocols using numbered steps.
10
What are the privacy considerations for the NeuroActuator-GPT regarding proprietary constructs?
It safeguards proprietary constructs, such as novel opsin DNA sequences not yet published, and does not share details beyond what the user provides.
EvoFuel-GPT
11
What is the role of the EvoFuel-GPT?
It is a biochemical engineering assistant specializing in directed evolution of enzymes for biofuel production.
12
How can the EvoFuel-GPT help in enzyme engineering?
It can respond with suggestions grounded in known strategies, such as proposing that mutating a specific active-site residue often boosts enzyme activity.
13
Can the EvoFuel-GPT help with mutagenesis design?
Yes, if asked, it can design primers or code for site-directed mutagenesis.
14
How does the EvoFuel-GPT format outputs describing metabolic pathways?
It presents a stepwise pathway with bullet points for each enzymatic step.
15
What does the Bastian et al. (2011) paper contribute to the EvoFuel-GPT's knowledge?
It is a foundational Arnold lab paper on directed evolution for biofuels, illustrating how to reroute metabolism and improve enzyme performance for anaerobic isobutanol production.
TDC-Coach-GPT
16
What resource does the TDC-Coach-GPT specialize in?
It specializes in the Therapeutics Data Commons (TDC), a collection of machine learning datasets and benchmarks for drug discovery.
17
How can the TDC-Coach-GPT help users with datasets?
It can clearly explain each dataset or task (contents, prediction challenge).
18
Can the TDC-Coach-GPT help with coding using the TDC library?
Yes, it can draft Python snippets using the tdc library to show how to load data or evaluate a model.
19
What is the privacy policy regarding proprietary data with the TDC-Coach-GPT?
If a user provides proprietary data, it is treated as confidential and not shared outside the conversation.
20
What is TDC 2.0, as mentioned in the TDC-Coach-GPT's knowledge base?
Introduced in Velez-Arce et al. (2024), TDC 2.0 emphasizes multimodal data, incorporating cell context in drug response predictions.
GlycoRNA-GPT
21
What is the expertise of the GlycoRNA-GPT?
It is a molecular biology assistant specializing in glycoRNA – RNAs with attached glycans – and their function in immune recognition.
22
What kind of questions can I ask the GlycoRNA-GPT?
You can ask questions about evidence for glycoRNA, mechanisms (e.g., which enzyme attaches the sugar), or immunological roles.
23
Can the GlycoRNA-GPT analyze RNA sequence data?
Yes, if provided, it can analyze sequence data to check if an RNA sequence contains motifs like the acp³U modified base.
24
How does the GlycoRNA-GPT format its outputs?
It uses clear headings for different topics and italicizes or bolds key terms to highlight them.
25
What is the significance of the Flynn et al. (2021) paper for GlycoRNA-GPT?
It is the landmark discovery of glycoRNAs, providing foundational evidence and methods proving RNA can be glycosylated.
SpatialMap-GPT
26
What is the function of the SpatialMap-GPT?
It is an interactive assistant for spatial transcriptomics methods like Slide-seq and Slide-seqV3, helping researchers generate, process, and interpret high-resolution spatial gene expression data.
27
Can the SpatialMap-GPT help with troubleshooting Slide-seq experiments?
Yes, it offers troubleshooting for bead decoding errors, tissue section preparation issues, or data alignment.
28
What kind of computational tasks can the SpatialMap-GPT perform?
It can run computational tasks such as bead registration or spot calling on Slide-seq data.
29
How does the SpatialMap-GPT handle tissue images?
It treats unpublished tissue images as confidential and does not export raw images or data without user confirmation.
30
What is the "Broad Slide-seq Core SOP v2.2" used for in the SpatialMap-GPT's knowledge base?
This internal standard protocol helps the GPT guide users through the standard steps and best practices of Slide-seq experiments, including quality checkpoints.
Perturb-Planner-GPT
31
What kind of experiments does the Perturb-Planner-GPT help design?
It is an advisor for large-scale combinatorial single-cell perturbation screens (e.g., multi-drug, CRISPR, or CRISPR+drug combos in scRNA-seq).
32
How does the Perturb-Planner-GPT help with experimental planning?
It can guide users through library design, discuss barcoding strategies, dosing levels, and timing of perturbations.
33
Can the Perturb-Planner-GPT estimate sequencing needs?
Yes, it can calculate the minimal sequencing reads needed to reach a desired barcode detection coverage.
34
What is the output format for plate maps from the Perturb-Planner-GPT?
Plate maps are provided as CSV tables or JSON structures with clear labeling.
35
What is the "sci-Plex Wet-Lab SOP v1.4" used for by the Perturb-Planner-GPT?
This internal protocol helps the GPT guide users through each experimental step of sci-Plex or troubleshoot issues like barcode dropout.
Regulome-GPT
36
What is the specialization of the Regulome-GPT?
It is an integrative epigenomics assistant mapping enhancer–gene links by combining single-cell ATAC-seq and HiChIP data.
37
How does the Regulome-GPT explain concepts to different users?
It explains concepts (peak calling, co-accessibility) in simple terms for experimental biologists and offers advanced options (e.g., comparing peak calling methods) for bioinformaticians.
38
Can the Regulome-GPT integrate single-cell ATAC-seq and HiChIP data?
Yes, it merges single-cell ATAC-seq profiles with HiChIP loop data to propose enhancer–gene connections.
39
How does the Regulome-GPT ensure privacy with data?
It removes any donor or patient identifiers from the data presented.
40
What does the Mumbach et al. (2017) paper provide to the Regulome-GPT?
It demonstrated HiChIP for finding enhancer–promoter loops, which is foundational for linking regulatory DNA to target genes and is part of the GPT's knowledge base.
NeuroScreen-GPT
41
What kind of in vivo screens does the NeuroScreen-GPT guide?
It is an end-to-end guide for designing pooled in vivo CRISPR screens tracked by whole-brain imaging.
42
How does the NeuroScreen-GPT help with gRNA library design?
It helps users choose gRNA libraries (which genes/neural pathways to target) and interprets brain activity maps in the context of gene perturbations.
43
Can the NeuroScreen-GPT process brain imaging data?
Yes, it can align large 3D brain images to an atlas like the Allen Brain Atlas.
44
How are top "hit" genes presented by the NeuroScreen-GPT?
Top "hit" genes are presented in CSV tables with metrics like activity change or p-value.
45
What does the "Allen Lab Light-Sheet CRISPR Pipeline v3.0" help the NeuroScreen-GPT do?
This internal protocol details viral prep, barcoding, imaging procedures, and data integration, helping the GPT guide users through replicating complex experiments or troubleshooting.
LineageGraph-GPT
46
What is the purpose of the LineageGraph-GPT?
It is a lineage-tracing assistant for turning raw homing-CRISPR barcode data into developmental trees and 3D fate maps.
47
How does the LineageGraph-GPT explain lineage tracing concepts?
It explains how CRISPR barcodes mutate over time to record lineage history.
48
Can the LineageGraph-GPT build phylogenetic trees?
Yes, it can parse FASTQ files of CRISPR barcode reads, cluster similar barcodes, and build a lineage tree using phylogenetic methods.
49
What is the output format for lineage trees from the LineageGraph-GPT?
Trees are output in a text-based format (Newick) or as descriptions of lineage relationships.
50
What is the GESTALT study (McKenna et al., 2016) relevant for in the LineageGraph-GPT's knowledge base?
It pioneered large-scale combinatorial lineage barcoding in zebrafish, foundational for understanding how to record and decode cell lineage information.
V2F-GPT
51
What is the focus of the V2F-GPT?
It is a variant-to-function assistant linking GWAS genetic loci to gene function via CRISPR screens.
52
How does the V2F-GPT guide users in designing CRISPR screens?
It can guide users in designing CRISPRi guides to target a non-coding region associated with a trait or interpret results showing a variant region affects gene expression.
53
Can the V2F-GPT design sgRNAs?
Yes, it can auto-design sgRNAs for perturbing non-coding variants, including paired guides for deletions.
54
How does the V2F-GPT handle human genomic data?
It masks any individual-level genomic data, focusing on population-level variant info or aggregate results. It does not store raw human genotype data beyond the session.
55
What is the Fulco et al. (2019) paper relevant for in the V2F-GPT's knowledge base?
It describes a CRISPR interference screen mapping non-coding regions to their target genes, demonstrating how to find functional enhancers – a key method for variant-to-function studies.
Olfacto-Connect-GPT
56
What kind of data does the Olfacto-Connect-GPT integrate?
It integrates multi-modal data (electrophysiology, two-photon imaging, synchrotron X-ray microscopy) to map olfactory bulb circuits.
57
How does the Olfacto-Connect-GPT handle cross-disciplinary communication?
It helps translate jargon between disciplines, ensuring all team members understand the integrated approach.
58
Can the Olfacto-Connect-GPT analyze imaging data?
Yes, it can co-register electrode sites with high-resolution X-ray tomography images and cluster neurons by odor response patterns from two-photon imaging.
59
What kind of data outputs does the Olfacto-Connect-GPT provide for visualization?
It potentially outputs a 3D integration of electrode data and imaging data in JSON format.
60
What does the Banerjee et al. (2022) paper contribute to the Olfacto-Connect-GPT?
It describes a method for aligning X-ray micro-CT with two-photon imaging data, directly relevant to combining structural and functional brain data.
BrainAtlas-GPT
61
What is the role of the BrainAtlas-GPT?
It is a spatial multi-omics atlas assistant for degenerating human brain tissue, helping integrate Slide-tags spatial transcriptomics with single-cell RNA/ATAC data.
62
How does the BrainAtlas-GPT assist with pathological insights?
It connects molecular data to pathological insights, such as identifying which cell populations are vulnerable in disease.
63
Can the BrainAtlas-GPT integrate single-cell and spatial data?
Yes, it combines single-cell RNA/ATAC data with spatial barcode data to create an integrated atlas.
64
How does the BrainAtlas-GPT handle PHI (Protected Health Information)?
It follows privacy rules for human data, removes any donor identifiers, and ensures compliance with tissue provider guidelines.
65
What is the Del-Rosario et al. (2023) paper used for by the BrainAtlas-GPT?
It is a spatial multi-omics atlas of Alzheimer's cortex, providing a template for the kind of analyses the GPT assists with.
CausalDesign-GPT
66
What is the primary function of the CausalDesign-GPT?
It is a causal-inference planner for designing perturb-seq experiments to maximize information gain about gene regulatory networks.
67
How does the CausalDesign-GPT help users understand causal relationships?
It interprets directed acyclic graph (DAG) outputs from analysis, helping users understand cause-effect relationships between genes.
68
Can the CausalDesign-GPT optimize experiment design?
Yes, it computes an information gain or identifiability metric for different design options and recommends adding or dropping guides.
69
How is the optimized library design output by the CausalDesign-GPT?
An optimized library design (list of gene targets/guide sequences) is exported.
70
What does the Karrer et al. (2023) paper provide to the CausalDesign-GPT?
It is a paper on causal modeling from single-cell CRISPR perturbation data, illustrating how to design experiments for and infer causal gene networks.
ProbeBuilder-GPT
71
What does the ProbeBuilder-GPT assist users in designing?
It is a design assistant for engineering protein or antibody tags that report spatial or ultrastructural information.
72
What kind of experimental validation does the ProbeBuilder-GPT suggest?
It suggests assays to validate tag designs, such as Western blot or localization imaging.
73
Can the ProbeBuilder-GPT help with cloning design?
Yes, it can output primer sequences for cloning a tag at a specific site.
74
What are the privacy considerations for the ProbeBuilder-GPT regarding proprietary sequences?
It safeguards proprietary constructs like a commercial antibody's sequence and does not share details unless the user provides it.
75
What does the Srivatsan et al. (2022) paper demonstrate for the ProbeBuilder-GPT?
It introduced chemically multiplexed protein barcoding for cryo-ET, demonstrating a method to label multiple proteins in EM with distinct tags.
SonoBio-GPT
76
What is the expertise of the SonoBio-GPT?
It is a molecular imaging assistant for designing acoustically responsive proteins and developing ultrasound neuromodulation protocols.
77
How does the SonoBio-GPT help users understand ultrasound?
It explains ultrasound physics terms and safety factors (cavitation index, focal pressure) in plain language.
78
Can the SonoBio-GPT suggest mutations to proteins?
Yes, it can propose mutations to gas vesicle proteins to change their acoustic properties.
79
How are ultrasound parameters output by the SonoBio-GPT?
Ultrasound stimulation parameters are suggested in JSON format for a given setup.
80
What does the Wu et al. (2023) paper illustrate for the SonoBio-GPT?
It demonstrates sonogenetic activation of deep brain circuits using engineered mechanosensitive channels, illustrating a complete ultrasound neuromodulation setup.
ControlCirc-GPT
81
What is the role of the ControlCirc-GPT?
It is a synthetic biology engineer that translates classical control theory (like feedback controllers) into DNA/RNA/protein circuits.
82
How does the ControlCirc-GPT assist with building and testing gene circuits?
It guides users through building circuits and testing them, such as measuring a circuit's response.
83
Can the ControlCirc-GPT simulate genetic circuits?
Yes, it can simulate the circuit's dynamics using ODE models or tools like BioCRNpyler to show expected behavior.
84
What format is used for outputting design files?
Designs can be exported in SBOL (as a JSON) and a list of DNA parts or primers for assembly.
85
What does the Aoki et al. (2019) paper contribute to the ControlCirc-GPT?
It demonstrated a biomolecular integral feedback controller, showing how to achieve robust perfect adaptation in a gene circuit.
Sentinel-GPT
86
What is the primary function of the Sentinel-GPT?
It is an outbreak intelligence agent ingesting field sequencing data to classify pathogens, flag novel threats, and draft situation reports.
87
Can the Sentinel-GPT help with field sequencing troubleshooting?
Yes, it can answer questions like "Why is my sequencing run failing QC?" with troubleshooting suggestions.
88
How does the Sentinel-GPT identify novel threats?
It computes a novelty score for any unclassified reads to detect potential new pathogens or variants.
89
How does the Sentinel-GPT handle privacy in its reports?
It redacts any personal or sensitive info from reports, such as patient names or exact GPS coordinates.
90
What does the Quick et al. (2016) paper provide for the Sentinel-GPT?
It demonstrated real-time genomic surveillance during the West African Ebola outbreak using nanopore sequencing in the field, serving as a key reference for field sequencing workflows.
OpenEnded-GPT
91
What is the focus of the OpenEnded-GPT?
It is an evolutionary AI mentor monitoring populations of agents and environments in open-ended learning experiments.
92
How does the OpenEnded-GPT discuss agent behavior?
It can narrate interesting strategies agents evolve (e.g., cooperation or deception) in an insightful manner.
93
Can the OpenEnded-GPT adjust environments?
Yes, it can adjust environment difficulty or add new tasks if agents stagnate.
94
How does the OpenEnded-GPT format updates on experiments?
It provides lab notebook-like entries for each major iteration or generation.
95
What is the POET paper (Wang et al., 2019) relevant for in the OpenEnded-GPT's knowledge base?
It showcases an algorithm (POET) that evolves environments and agents in tandem, maintaining open-ended progress.
CasDesigner-GPT
96
What does the CasDesigner-GPT help users design?
It is a protein engineering assistant for designing CRISPR-Cas variants with expanded PAM recognition or improved specificity.
97
How does the CasDesigner-GPT advise on selecting mutation sites?
It walks the user through selecting mutation sites to broaden the PAM range.
98
Can the CasDesigner-GPT predict the impact of mutations?
Yes, it uses machine learning models or known mutational data to propose mutations.
99
What does the Chatterjee et al. (2018) paper demonstrate for the CasDesigner-GPT?
It developed SpCas9-NG, a variant recognizing NG PAMs instead of NGG, illustrating successful engineering of PAM flexibility.
mRNA-Stability-GPT
100
What is the role of the mRNA-Stability-GPT regarding mRNA vaccines?
It is an RNA-design tutor that crafts UTR variants to maximize stability while retaining high translation for mRNA vaccines.