Biological Engineering Application This project proposes the development of an intestinal spheroid culture platform derived from cell lines (e.g., Caco-2 spheroid or organoid-like cultures), combined with multi-omics profiling (transcriptomics, proteomics, and metabolomics) computational modeling using systems biology and machine-learning approaches. The platform is intended to support research on drug absorption, inflammatory bowel disease (IBD) diagnostics, and predictive analysis of treatment outcomes. Initially, the system will be used to generate hypotheses from experimental data, with the long-term goal of becoming a predictive research tool.
1-Benchling-in-silico-gel-art Using Benchling.com, Lambda DNA, Paul Vanouse’s Latent Figure Protocol artworks, and Ronan’s website as references, and incorporating creative design principles, simulations of restriction enzyme digests of the Lambda genome were performed using EcoRI, HindIII, BamHI, KpnI, EcoRV, SacI, and SalI:
Figure 1. Virtual restriction digest of Lambda DNA.
Once the banding patterns were characterized, images inspired by the previously mentioned works were created:
Assignment: Python Script for Opentrons Artwork Based on the Lissajous function, the figure to be created on the agar will be the following:
Post-Lab Questions — DUE BY START OF FEB 24 LECTURE Paper: Automation of biochemical assays using an open-sourced, inexpensive robotic liquid handler Moukarzel et al. 2024
Subsections of Homeworktype: chapter
Week 1 HW: Principles and Practices
[Spheroid Cell Culture]
1. Biological Engineering Application
This project proposes the development of an intestinal spheroid culture platform derived from cell lines (e.g., Caco-2 spheroid or organoid-like cultures), combined with multi-omics profiling (transcriptomics, proteomics, and metabolomics) computational modeling using systems biology and machine-learning approaches. The platform is intended to support research on drug absorption, inflammatory bowel disease (IBD) diagnostics, and predictive analysis of treatment outcomes. Initially, the system will be used to generate hypotheses from experimental data, with the long-term goal of becoming a predictive research tool.
Platform Workflow
3D Intestinal Model Generation: Establishment of Caco-2–derived 3D epithelial cultures to model intestinal barrier function.
Experimental Perturbation: Exposure of cultures to inflammatory signals, drug compounds, or microbiota-related metabolites. Multi-Omics Acquisition: Collection of transcriptomic, proteomic, and metabolomic data to capture cellular responses.
Data Processing and Integration: Quality control, normalization, and integration of omics datasets using reproducible bioinformatics pipelines.
Computational Modeling: Application of systems biology and machine-learning approaches to identify patterns and generate hypotheses.
Validation and Iteration: Experimental validation of model predictions through iterative testing.
2. Governance Framework
Governance Objectives
The project integrates governance principles to ensure safe, transparent, and equitable use of the technology.
Scientific and Clinical Safety
Implement staged validation protocols before diagnostic use.
Establish quality-control standards for omics data.
Limit early platform use to research contexts.
Document uncertainty in predictive models.
Biological Data Protection
Anonymize patient-derived data.
Comply with research ethics and data protection regulations.
Implement controlled access to datasets and software.
Maintain traceability of samples and analyses.
Responsible Use of Predictive Models
Design software as a research-support tool.
Include confidence and uncertainty metrics in predictions.
Validate models with independent datasets.
Avoid automated decision-making without human supervision.
Equity and Access
Promote open-source computational tools.
Design scalable experimental protocols.
Encourage collaboration with public institutions.
Document methodologies for technology transfer.
3. Governance Actions
Stage-based validation requirement: Restrict initial platform use to research applications until validation standards are met. In the early stages, use cell lines as a working model (3D spheroids/organoids).
Controlled access data management: Use public databases to triangulate working hypotheses. Implement anonymized datasets with institutional oversight and traceability.
Transparent computational workflows: Share bioinformatics processes and documentation through reproducible research practices.
Prioritization Strategy
The project prioritizes a combination of staged validation protocols and open, reproducible computational standards. These actions balance scientific safety with research feasibility and transparency. Controlled-access data infrastructure will be implemented progressively when human biological samples are incorporated.
4. Rating of governance actions
The following table summarizes the evaluation of governance options according to course criteria.
Based on the scoring above, the priority would be a combination of Option 1 (staged validation requirement) and Option 3 (open and reproducible computational standards). Together, these actions balance scientific safety with research feasibility. Validation protocols reduce the risk of incorrect interpretation or premature diagnostic use, while reproducible computational workflows promote transparency, collaboration, and constructive scientific applications without significantly increasing costs.
The project will follow validation-driven research practices, responsible data governance, and open computational workflows. Periodic ethical evaluation will accompany platform development to identify risks and support responsible translation into diagnostic or predictive applications.
Ethical Reflection and Protocol Standardization
To improve reproducibility and reliability, the project will:
Implement standard operating procedures (SOPs).
Validate and benchmark protocols across experiments.
Use shared documentation and version control for methods.
Assignment (Week 2 Lecture Prep) — DUE BY START OF FEB 10 LECTURE
Professor Jacobson’s homework questions:
Nature’s machinery for copying DNA is called polymerase. What is the polymerase error rate? How does this compare to the length of the human genome? How does biology address this discrepancy?
DNA polymerase with proofreading activity (3′-5′ exonuclease) has an approximate error rate of 1 in 10⁶ nucleotides incorporated.
The human genome has approximately 3.2 × 10⁹ base pairs, so if only polymerase fidelity existed, thousands of errors would be introduced per complete genome replication.
Biology resolves this discrepancy through multiple levels of error correction, for example:
polymerase proofreading
DNA repair systems (e.g., mismatch repair such as MutS)
redundancy and robustness of the biological system
Together, these mechanisms reduce the effective error rate to levels compatible with genome stability.
How many different ways are there to code (DNA nucleotide code) for an average human protein? In practice, what are some of the reasons why all these different codes fail to encode the protein of interest?
The genetic code has 64 codons for 20 amino acids, which leads to genetic code degeneracy. According to the slides, an average human protein has approximately 1036 bp, or about 345 amino acids.
If each amino acid can be encoded by an average of 3 codons, the number of possible sequences would be approximately 3^345, representing an extremely large number of possible DNA sequences that code for the same protein.
Why not all of these sequences work in practice. Many variants don’t work well due to biological and physical constraints, for example:
codon bias and translation efficiency (different codons for the same amino acid)
GC content and DNA stability
DNA/RNA secondary structures
unwanted regulatory signals
repeats or sequences that are difficult to synthesize
mRNA stability
In other words, even though the genetic code is redundant, not all equivalent sequences are functionally equivalent.
Dr. LeProust’s Homework Questions:
What is the most commonly used method for oligonucleotide synthesis?
Why is it difficult to produce oligonucleotides longer than 200 nucleotides by direct synthesis?
Why can’t a 2000 bp gene be created by direct oligonucleotide synthesis?
The most commonly used method for oligonucleotide synthesis is solid-phase phosphoramide synthesis. This process occurs cyclically, adding one nucleotide at a time, and each step has an efficiency slightly less than 100%. Due to this imperfect efficiency in each cycle, the probability of accumulated errors increases with the sequence length, making it difficult to produce oligonucleotides longer than approximately 200 nucleotides by direct synthesis. For the same reason, it is not possible to directly synthesize a 2000 base pair gene as a single oligonucleotide, as the accumulation of errors and truncated products would be too high. In practice, long genes are constructed by assembling multiple shorter oligos using molecular assembly methods (e.g., PCR assembly or Gibson assembly).
George Church’s Homework Question:
Choose ONE of the following three questions to answer; and cite any AI prompts or paper citations used.
[Using Google Slide #4 and Professor Church] What are the 10 essential amino acids in all animals? And how does this affect your view of the “Lysine Contingency”?
The essential amino acids in animals include: histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine (and in some contexts, arginine). These amino acids must be obtained from the diet because animals cannot synthesize them. In relation to the “Lysine Contingency,” lysine becomes a critical point in bioengineering because its metabolic availability can be used as a control mechanism or biological dependency in synthetic systems. This illustrates how natural metabolic constraints can be exploited as biocontainment or functional control strategies in synthetic biology.
[Given slides #2 and 4 (AA:NA and NA:NA codes)] What code would you suggest for AA:AA interactions?
A possible code for AA:AA interactions could be based on physicochemical complementarity (e.g., charge, hydrophobicity, and size), analogous to how NA:NA interactions rely on base pairing and AA:NA interactions rely on codon recognition. This is because amino acid–amino acid interactions are primarily determined by chemical and structural complementarity rather than a fixed symbolic code, unlike nucleic acid base pairing.
Week 2 HW: DNA Read, Write, & Edit
1-Benchling-in-silico-gel-art
Using Benchling.com, Lambda DNA, Paul Vanouse’s Latent Figure Protocol artworks, and Ronan’s website as references, and incorporating creative design principles, simulations of restriction enzyme digests of the Lambda genome were performed using EcoRI, HindIII, BamHI, KpnI, EcoRV, SacI, and SalI:
Figure 1. Virtual restriction digest of Lambda DNA.
Once the banding patterns were characterized, images inspired by the previously mentioned works were created:
Part 2: Gel Art - Restriction Digests and Gel Electrophoresis
Part 3: DNA Design Challenge
3.1. Choosing a protein:
Protein Chosen: INS Homo sapiens (Insulin): This is a known protein, short sequence, and without stop codons:
AAP35454.1 insulin [Homo sapiens] MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFYTPKTRREAEDLQVGQVELGGGPGAGSLQPLALEGSLQKRGIVEQCCTSICSLYQLENYCN
Obtaining the DNA sequence in tblastn:
In tblastn with the Human INS sequence (AAP35454.1), the search was on for the DNA sequence with the highest identity.
Figure 2. tblast of the selected sequence.
Once the result is obtained, by entering the sequence ID (AB587580.1) the gene sequence is obtained:
Figure 3. DNA selected sequence.
BT006808.1 Homo sapiens insulin mRNA, complete cds
ATGGCCCTGTGGATGCGCCTCCTGCCCCTGCTGGCGCTGCTGGCCCTCTGGGGACCTGACCCAGCCGCAGCCTTTGTGAACCAACACCTGTGCGGCTCACACCTGGTGGAAGCTCTCTACCTAGTGTGCGGGGAACGAGGCTTCTTCTACACACCCAAGACCCGCCGGGAGGCAGAGGACCTGCAGGTGGGGCAGGTGGAGCTGGGCGGGGGCCCTGGTGCAGGCAGCCTGCAGCCCTTGGCCCTGGAGGGGTCCCTGCAGAAGCGTGGCATTGTGGAACAATGCTGTACCAGCATCTGCTCCCTCTACCAGCTGGAGAACTACTGCAACTAG
3.2. Reverse translation: Protein sequence (amino acids) to DNA sequence (nucleotides)
The central dogma, discussed and recited in class, describes the process by which the DNA sequence is transcribed and translated into protein. The central dogma provides the framework for working in reverse from a given protein sequence and inferring the DNA sequence from which it is derived. Using one of the tools discussed in class, the NCBI (tblastn) or online tools (search “reverse translation tools” on Google), determine the nucleotide sequence that corresponds to the protein sequence you chose earlier.
3.3. Codon optimization
Codon optimization: on the web site NovoProLabs modifying the nucleotide sequence of a gene to maximize protein production in a specific organism by adapting codon usage to the host’s codon bias. Gene expression in a different organism may require adjusting codon usage to match the host’s translational preferences.
Restriction enzymes EcoRI and XhoI are not involved in the codon optimization itself, but are used to facilitate cloning of the optimized gene into an expression vector. During sequence design, the optimization tool adds the specific recognition sites of these enzymes to the ends of the gene, so that, by digesting both the insert and the plasmid with the same enzymes, compatible cohesive ends are generated that allow the gene to be inserted in the correct orientation within the vector. Instead, the changes observed in Relative Adaptiveness and GC content come exclusively from the codon optimization process, which modifies the nucleotide sequence without altering the amino acid sequence to improve translation efficiency in the host organism.
Codon optimization of the insulin coding sequence for E. coli expression increased the Codon Adaptation Index (CAI) from 0.48 to 0.90, indicating improved compatibility with the host’s codon usage preferences and a higher expected translation efficiency. At the same time, GC content was adjusted from 64.56% to 60.00%, bringing it closer to a balanced range that can improve mRNA stability and transcriptional performance. Overall, the optimization modifies the nucleotide sequence without changing the amino acid sequence, with the goal of enhancing recombinant protein expression in the bacterial system.
3.4 Optimized protein transcription and translation
To produce this protein from DNA, the optimized gene can be inserted into a recombinant expression vector and introduced into a host such as *E. coli *using recombinant DNA technology and heterologous expression systems. Inside the cell, the DNA sequence is first transcribed into mRNA by RNA polymerase, and the mRNA is then translated by ribosomes into the protein according to the genetic code. Alternatively, the same DNA template can be used in cell-free expression systems (in vitro transcription - translation, IVTT), which contain purified enzymes, ribosomes, tRNAs, and energy sources that allow transcription and translation to occur outside living cells. Both approaches rely on the central dogma of molecular biology- DNA => RNA => protein - but differ in whether protein production occurs in living cells or in a controlled biochemical system.
3.5 Alternative splicing
A single gene can produce multiple proteins at the transcriptional level mainly through alternative splicing (alternative splicing). During pre-mRNA processing, different exons can be included or excluded in different combinations, generating multiple mature mRNAs from the same gene. Each of these mRNAs can be translated into different protein isoforms, with variations in their structure and function. Furthermore, mechanisms such as alternative promoters or alternative polyadenylation sites can also produce different transcripts from the same gene locus.
Part 4: Prepare a Twist DNA Synthesis Order
4.1. Create a Twist account, and Benchling account
4.2. Build Your DNA Insert Sequence
In Benchling, select New DNA/RNA sequence. Give your insert sequence a name and select DNA with a Linear topology (this is a linear sequence that will be inserted into a circular backbone vector of our choosing).
The sequence of plasmid PP533546.1 was downloaded from GenBank and used as cloning backbone. Then, a synthetic INS coding sequence from Homo sapiens was designed in Benchling. After preparing both sequences, virtual DNA assembly was performed to insert the INS construct into the plasmid backbone, generating a recombinant circular plasmid containing the expression cassette (promoter, RBS, INS CDS, His tag, and terminator). This in-silico cloning step allowed visualization and verification of the final construct.
(i) What DNA would you want to sequence (e.g., read) and why? This could be DNA related to human health (e.g. genes related to disease research), environmental monitoring (e.g., sewage waste water, biodiversity analysis), and beyond (e.g. DNA data storage, biobank).
Sanger sequencing is a first-generation DNA sequencing method based on chain termination. During DNA synthesis, modified nucleotides (dideoxynucleotides, ddNTPs) are incorporated randomly, stopping elongation. The resulting fragments of different lengths are separated by capillary electrophoresis, and the sequence is read from fluorescent labels. It is highly accurate and ideal for validating plasmids or specific genes, but it is low-throughput and typically limited to ~700 - 1000 bp per read.
Nanopore sequencing is a third-generation method that sequences DNA in real time by passing single DNA molecules through a biological nanopore. As nucleotides move through the pore, they disrupt an ionic current in characteristic ways, allowing base identification. It can generate very long reads (even entire plasmids or genomes in one fragment) and works without PCR amplification, but raw accuracy can be slightly lower than Sanger, though it has improved significantly.
An example of next-generation sequencing (NGS) is Illumina sequencing. It uses massively parallel sequencing by synthesis, where millions of DNA fragments are immobilized on a flow cell and sequenced simultaneously through cyclic incorporation of fluorescently labeled nucleotides. This approach provides extremely high throughput and is commonly used for whole-genome sequencing, transcriptomics (RNA-seq), and large-scale variant analysis.
(ii) What technology or technologies would you use to perform your DNA sequencing and why?
Genes of the gut microbiota. To sequence gut microbiota genes, the most widely used and appropriate technology would be second-generation sequencing (NGS), especially platforms like Illumina. Third-generation sequencing, such as Oxford Nanopore Technologies, could also be considered, depending on the objective (taxonomic resolution, or complete assembly).
Responda también las siguientes preguntas:
Is your method first, second, or third generation, or something else? What does that mean?
Recommended primary technology: Illumina (NGS). What generation is it? It’s second generation. It’s characterized by performing millions of reads in parallel (massively parallel sequencing) with short fragments and high accuracy.
For gut microbiota studies (e.g., 16S rRNA gene sequencing or metagenomics), Illumina is ideal due to its high accuracy, low cost per sample, high sequencing depth, and excellent bioinformatics support. Nanopore would be useful if you are looking for long reads or whole genome assembly.
What is your opinion? How do you prepare your information (e.g., fragmentation, adapter ligation, PCR)? List the essential steps.
Sample Preparation (e.g., 16S rRNA or metagenomics). Essential Steps:
DNA extraction from the fecal sample.
Fragmentation (if metagenomics; not always necessary for 16S rRNA).
PCR amplification
For 16S rRNA: amplification of variable regions (V3–V4).
Ligation of adapters and indices (barcodes).
Purification and quantification.
Loading into the flow cell of the sequencer.
What are the essential steps of the sequencing technology you have chosen and how does it decode the bases of your DNA sample (base calling)?
How does Illumina technology work?
DNA with adapters is attached to a flow cell.
A cluster is generated by bridge amplification.
Reversibly fluorescent terminator nucleotides are incorporated.
Each cycle adds one base.
A camera detects the fluorescence.
The terminator is removed, and the cycle repeats.
How are the bases decoded? Each base (A, T, C, G) emits a different fluorescence. The system records the optical signal and converts it into a digital sequence (base calling).
What is the outcome of the chosen sequencing technology?
The end result is millions of short reads (FASTQ files). Each read includes base sequence and quality scores (Phred score). Subsequently, bioinformatics analysis, taxonomic identification, bacterial abundance profiling, and alpha and beta diversity are performed.
5.2 DNA writing
What DNA would you like to synthesize (e.g., write) and why?
Genes that are expressed in inflammatory settings. Genes that are artificially expressed so that in IBD contexts, they are expressed in cell cultures, spheroids, or organoids.
It should serve as a reliable indicator, expressed through the presence of butyrate, SCFAs, or various bacterial metabolites, or oxidative stress scenarios. It should be activated by the presence of a common marker metabolite.
A promoter sensitive to an inflammatory or metabolic signal controls the expression of a reporter gene (e.g., GFP, luciferase, mCherry, etc.). When the stimulus appears (butyrate, ROS, NF-κB, etc.), the promoter is activated and the reporter gene is expressed.
(ii) What technology or technologies would you use to perform this DNA synthesis and why?
To design and validate a promoter sensitive to inflammatory or metabolic signals, it is first necessary to identify endogenous regulatory regions that respond to the stimulus of interest using NGS technologies such as RNA-seq to detect stimulus-induced genes, and ATAC-seq or ChIP-seq to map open chromatin regions or binding sites of specific transcription factors (e.g., NF-κB). Complementarily, techniques such as STARR-seq allow for the functional evaluation of promoter activity across thousands of sequences simultaneously, identifying those that actually activate transcription under the stimulus. By combining these data, synthetic promoters can be designed to drive the specific and quantifiable expression of a reporter gene (GFP, luciferase, mCherry) in response to the desired signal.
Please also answer the following questions:
What are the essential steps of the sequencing methods you have chosen?
What are the limitations of your sequencing method (if any) in terms of speed, accuracy, and scalability?
The essential steps of RNA-seq are: (1) isolation of total RNA or messenger RNA (mRNA) from the samples of interest; (2) fragmentation and library preparation, which includes conversion to cDNA, ligation of adapters, and sometimes size selection; (3) sequencing of the libraries on an NGS platform to obtain short reads; (4) data processing, which involves quality control, filtering, alignment of the reads to the reference genome or transcriptome, and quantification of gene expression; and (5) differential analysis to identify genes whose expression level changes between conditions, followed by functional annotation and pathway analysis.
The main limitations of RNA-seq in terms of speed, accuracy, and scalability are:
Speed: Library preparation and sequencing can take anywhere from several hours to several days, especially if biological replicates and multiple conditions are required; bioinformatics analyses can also be slow if the datasets are large.
Accuracy: Quantification can be affected by amplification bias, reverse transcription efficiency, fragmentation differences, or ambiguous mapping of reads to homologous genes or isoforms; genes with low expression are difficult to detect reliably.
Scalability: Processing many samples simultaneously increases costs and complexity, and storing and analyzing large volumes of data requires robust computational infrastructure; furthermore, high-resolution methods such as single-cell RNA-seq exponentially increase the amount of data and the complexity of the analysis.
5.3 DNA Editing
(i) Which DNA would you like to edit and why?
CRISPR editing. I believe it’s a technique with a very promising future.
(ii) What technology or technologies would you use to perform these DNA edits and why?
Also answer the following questions:
How does your preferred technology edit DNA? What are the essential steps?
CRISPR edits DNA using a Cas enzyme guided by a guide RNA that recognizes a specific sequence in the genome. Cas cuts the DNA at that point, and then the cell repairs the break, which can result in insertions, deletions, or allow the incorporation of a new sequence if a repair template is provided.
What preparation is needed (e.g., design steps) and what information (e.g., DNA template, enzymes, plasmids, primers, guides, cells) is required for editing?
A guide RNA that recognizes the target sequence needs to be designed, and, if insertion is desired, a DNA template with the sequence to be incorporated. Information and materials include: Cas9 enzyme (or similar), plasmids or delivery vectors, guide RNA, verification primers, the repair template if applicable, and the cells to be edited.
What are the limitations of your editing methods (if any) in terms of efficiency or accuracy?
CRISPR has limitations in efficiency, since not all cells correctly receive the editing machinery or repair DNA in the desired way, and in precision, because off-target effects or unforeseen insertions/deletions can occur at the cutting site. Furthermore, efficiency and precision depend on cell type, guide RNA design, and genomic context.
title: ‘Week 3 HW: Lab Automation’
weight: 30
Assignment: Python Script for Opentrons Artwork
Based on the Lissajous function, the figure to be created on the agar will be the following:
Post-Lab Questions — DUE BY START OF FEB 24 LECTURE
Paper: Automation of biochemical assays using an open-sourced, inexpensive robotic liquid handler
The study aimed to evaluate the feasibility of using an open-sourced, low-cost robotic liquid handler—specifically the Opentrons OT-2—for automating biochemical assays that are traditionally run on expensive industrial liquid handling platforms. High-throughput screening is a core process in pharmaceutical development, but the cost and training requirements of conventional robotic systems can be prohibitive. The authors set out to determine whether a lightweight, Python-programmable robot could perform common assay workflows with sufficient precision and reliability to be useful in early-stage assay development and method transfer.
To test this, the team programmed the OT-2 to perform two standard biochemical assays—PicoGreen for DNA quantification and Bradford for protein concentration—using custom Python protocols that controlled pipetting and reagent transfers across microplates. These automated workflows were run repeatedly and compared to runs on a more expensive Tecan EVO liquid handler to benchmark performance. The study measured pipetting accuracy, variability, and overall assay consistency to assess how well the OT-2 handled the tasks relative to the industrial system.
The results showed that the OT-2 delivered accurate pipetting with low covariance across replicates, demonstrating performance close to that of the Tecan EVO despite its substantially lower cost and simpler hardware. Although limitations such as the absence of a crash detection system and a relatively small deck space were noted, the robot’s affordability and flexibility were highlighted as significant advantages. The authors concluded that the OT-2 represents a cost-effective, medium-throughput automation solution well suited for early-stage assay development and method transfer without requiring large capital investments.
Week 3 HW: Lab automation
Assignment: Python Script for Opentrons Artwork
Based on the Lissajous function, the figure to be created on the agar will be the following:
Post-Lab Questions — DUE BY START OF FEB 24 LECTURE
Paper: Automation of biochemical assays using an open-sourced, inexpensive robotic liquid handler
The study aimed to evaluate the feasibility of using an open-sourced, low-cost robotic liquid handler—specifically the Opentrons OT-2—for automating biochemical assays that are traditionally run on expensive industrial liquid handling platforms. High-throughput screening is a core process in pharmaceutical development, but the cost and training requirements of conventional robotic systems can be prohibitive. The authors set out to determine whether a lightweight, Python-programmable robot could perform common assay workflows with sufficient precision and reliability to be useful in early-stage assay development and method transfer.
To test this, the team programmed the OT-2 to perform two standard biochemical assays—PicoGreen for DNA quantification and Bradford for protein concentration—using custom Python protocols that controlled pipetting and reagent transfers across microplates. These automated workflows were run repeatedly and compared to runs on a more expensive Tecan EVO liquid handler to benchmark performance. The study measured pipetting accuracy, variability, and overall assay consistency to assess how well the OT-2 handled the tasks relative to the industrial system.
The results showed that the OT-2 delivered accurate pipetting with low covariance across replicates, demonstrating performance close to that of the Tecan EVO despite its substantially lower cost and simpler hardware. Although limitations such as the absence of a crash detection system and a relatively small deck space were noted, the robot’s affordability and flexibility were highlighted as significant advantages. The authors concluded that the OT-2 represents a cost-effective, medium-throughput automation solution well suited for early-stage assay development and method transfer without requiring large capital investments.
Final Project Description – Automation of ABC Transporter Uptake and Efflux Assays in Intestinal Organoids
To develop a semi-automated workflow for ABC transporter uptake and efflux assays using intestinal spheroid and organoid cultures in a 6, 12 and 96-well plate format. The objective is to improve reproducibility, throughput, and quantitative accuracy while reducing manual variability in washing, incubation timing, and sample collection steps.
1- Automation:
A- Cell Culture:
a. Cell counter (density according to culture type). Create a script or application to count the number of cells. b. Plate treatment. Drying and seeding. Opetrons
B- Automation of Treatments:
a. Cell culture media changes b. Washes with buffer of cultures in pretreatments c. Media changes and special media
C- ANT (total nucleic acids) extraction
D- Bioinformatics:
a. Census of different crop variables and of diagnostic interest. Ej.: pH, CO2, Temperature
E- Multiomics
2. Liquid Handling Automation
To design a pipetting workflow compatible with a benchtop liquid handler (e.g., Opentrons-like platform). The automated protocol will: 1- Remove culture medium 2- Wash wells with PBS (1–2 cycles, optimized). 3- Add loading medium with defined metabolite concentrations. 4- Incubate for a programmable time. 5- Remove loading medium. 6- Perform PBS washes. 7- Add HBSS efflux buffer. 8- Incubate for defined time intervals. 9- Transfer efflux supernatant to a secondary 96-well plate or tubes. 10- Add diluted Triton for cell lysis. Timing precision will be critical, especially for efflux kinetics.
Considerations:
Automatic mapping of conditions per well (using an imported CSV file). Differential control by column (e.g., column 1–3 control, 4–6 MK-571). Automatic metadata recording (plate ID, date, batch). Kinetic analysis at multiple intervals (e.g., collect at 5, 10, 15 min).
B. Example Pseudocode (Conceptual Workflow)
Example Opentrons Protocol – ABC Transporter Efflux Assay (96-well format)
