Session 1: Foundations of Discovery Omics: Proteomics, Metabolomics & Biomarker Discovery Workshop

Summary

Workshop introduction and platform access

The workshop organizer introduced the program as part of a clinical proteomics and metabolomics series. The organization offers courses, workshops, internships, research programs, and mentor-guided training, with an emphasis on hands-on learning.

Participants were shown how to access workshop materials:

• Create an account on the LogicLearn portal using an email and password or a linked account.
• Send the organizer the email used to register, or share a profile link, so workshop access can be activated.
• Check that the profile name is spelled and capitalized correctly; it may appear on the workshop certificate.
• Open the Programs tab, select the workshop page, and bookmark it for easy access.
• Find session recordings and resources—such as datasets, databases, and reading materials—under Syllabus / Interactive Sessions. These are uploaded after each session.

Proteomics fundamentals

The guest lecturer focused primarily on mass-spectrometry-based bottom-up proteomics. In this approach, proteins are digested into peptides before analysis. This differs from top-down proteomics, which examines intact proteins.

Proteomics is challenging because:

• One gene can produce multiple protein forms through alternative splicing and post-translational modifications.
• Protein abundance varies enormously. This difference can be especially large in complex samples such as blood, where abundant proteins may obscure low-abundance ones.
• Mass spectrometers have finite detection ranges, so sample preparation and experimental design are crucial.
• Instrument technology has improved, but sample preparation remains a major determinant of data quality.

Bottom-up proteomics workflow

The lecturer outlined a typical workflow:

1. Extract proteins from cells or tissue by lysing the sample, denaturing and solubilizing proteins, and controlling protease and nuclease activity.
2. Reduce and alkylate proteins to unfold them and prevent disulfide bonds from reforming, making proteins more accessible to digestion.
3. Digest proteins into peptides, commonly with trypsin, which cleaves at lysine and arginine residues. Digestion conditions—including pH and salt concentration—affect the result.
4. Clean up the sample to remove salts, detergents, and other substances that can interfere with mass spectrometry.
5. Enrich or deplete components when needed; for example, remove highly abundant plasma proteins to improve detection of less abundant proteins.
6. Separate peptides by liquid chromatography (LC) before they enter the mass spectrometer.
7. Ionize and analyze the peptides, then use software to identify peptides and infer the proteins present.

The lecturer stressed that buffers and detergents must be compatible with mass spectrometry or removed before injection. Urea can cause unwanted carbamylation under heating conditions, so its use and storage require care. Clean handling—such as wearing fresh gloves and avoiding keratin contamination from skin or hair—also matters.

Quantification strategies

Three broad approaches were discussed:

• Label-free quantification: Does not require labeling reagents and is relatively inexpensive per sample, but each sample is typically run separately.
• Isobaric labeling: Samples receive distinct tags and are combined before analysis. This can reduce run-to-run variation, but the tags and related workflows can be costly.
• Metabolic labeling: Uses heavy isotopes in cell-culture media. It is useful for cultured cells but is generally not suited to complex samples such as plasma or tissue.

The choice depends on the study design, sample type, budget, and number of conditions.

Mass spectrometry and chromatography

A mass spectrometer measures the mass-to-charge ratio (m/z) of ions. Its main components are:

• An ion source, where analytes are charged.
• A mass analyzer, which separates or filters ions by m/z.
• A detector, which records the signal.

The instrument operates under high vacuum. Electrospray ionization (ESI) is commonly coupled with LC for proteomics. MALDI is another ionization method and can tolerate higher salt concentrations, but generally produces singly charged ions.

LC helps separate peptides before analysis. Nanoflow LC can provide high sensitivity with small sample amounts, but may be prone to clogging and other operational problems. Microflow LC offers an intermediate option. Retention-time reproducibility matters because shifts between runs can complicate peptide matching. Poor separation can also allow abundant peptides to mask less abundant ones.

Mass analyzers discussed included quadrupoles, ion traps, time-of-flight (TOF) analyzers, and Orbitraps. Quadrupoles can act as mass filters. Triple-quadrupole instruments are widely used for targeted quantification, while TOF- and Orbitrap-based instruments are common in discovery work. Fragmentation—often collision-induced dissociation or higher-energy collisional dissociation—produces peptide fragments used for sequence identification. Alternative fragmentation methods can be useful for studying certain modifications.

Data acquisition and analysis considerations

Two acquisition approaches were contrasted:

• Data-dependent acquisition (DDA): The instrument selects and fragments the most abundant ions in a given sampling cycle. This can produce high-quality spectra, but low-abundance peptides may be missed.
• Data-independent acquisition (DIA): The instrument fragments ions across broader m/z windows, aiming to capture more peptides, including lower-abundance ones. DIA produces larger, more complex datasets, making analysis and error control important. DDA data can also be used to build libraries for DIA analysis.

The lecturer noted that downstream analysis must account for:

• Normalization and possible changes in instrument response between runs.
• Missing values, which may reflect either low abundance or technical nondetection.
• Batch effects, especially in clinical studies. Controls or quality-control samples distributed through a run can help monitor instrument performance.

Biomarker discovery and clinical translation

The lecturer described a general biomarker-development path:

1. Discovery: Use an exploratory experiment to identify candidate proteins and generate hypotheses.
2. Qualification or verification: Test candidates in a larger cohort, often with targeted methods such as PRM or MRM.
3. Validation: Assess whether the findings hold in larger, potentially multi-site studies.
4. Assay development and regulatory review: Develop a reproducible assay and evaluate whether it is suitable for clinical use.

Many reported biomarkers do not reach clinical practice. The lecturer attributed this largely to study-design and validation problems rather than to mass spectrometers alone. Issues include insufficient sample sizes, poorly matched control groups, differences in sample collection or storage, batch effects, weak statistical models, and biomarkers that are not specific to the disease of interest. Clinical translation is also limited by the time and complexity of proteomics workflows and regulatory requirements.

Targeted proteomics

Targeted approaches were presented as hypothesis-driven methods for testing a focused set of candidate proteins. They can be used to verify findings from discovery studies and are conceptually similar to testing selected targets with an antibody-based assay, but use mass spectrometry to detect peptides.

• SRM and MRM refer to closely related approaches for monitoring selected peptide transitions.
• PRM uses a high-resolution analyzer to measure product ions.
• Targeted methods require substantial optimization, including choosing suitable peptides, establishing retention times, selecting transitions, and tuning fragmentation energy.
• A peptide that is detectable is not automatically suitable for quantification. The lecturer distinguished proteotypic peptides from quantotypic peptides, which more reliably reflect the amount of the protein.
• Candidate targets can come from discovery data, literature, gene-expression or protein-interaction data, and public resources such as PeptideAtlas or PRIDE.

Practical advice from the discussion

The lecturer recommended planning experiments carefully before beginning, using clean and consistently prepared reagents, checking buffer pH, and avoiding contamination. For long-term instrument quality control, a lab can prepare a consistent peptide digest, aliquot it, and use it across runs. For targeted experiments, a heavy-labeled version of the peptide of interest can serve as a control.

The lecturer also noted that solvent choices depend on the application; common LC solvents for proteomics include water and acetonitrile with formic acid.

Speakers and sources featured

• Workshop organizer/presenter: Introduced as Jvasha; later addressed as Jibisha in the subtitles. She presented the program and portal instructions.
• Guest lecturer: Name is unclear in the autogenerated subtitles. He described himself as a postdoctoral research fellow at Boston Children’s Hospital and Harvard Medical School, with a PhD from the proteomics lab at IIT Bombay and approximately 12 years of mass-spectrometry and proteomics experience.
• Participants who introduced themselves or asked questions: Anamika Baki; Dr. Pragas (name spelling uncertain); Gemma; Dr. Snehi; Sheet (name unclear); Carmen; Dupati; Shini; Mel Peña (also addressed as Kamari in the subtitles); Niba/Nibita Mitra; Minu; and Akanga (audio unclear).
• People mentioned but not speaking in this session: Dr. Taru Biswas, Deep, and Ankit were referenced in connection with later workshop sessions.