Project Details

What You Will Do

Data Acquisition, Governance & Reproducible Setup

Complete data-access applications, build a clinical metadata dictionary and sample-level audit trail, set up a version-controlled R/Python environment, and fix inclusion, exclusion and missing-data rules before any analysis begins.

Raw NGS Processing & Quality Control

FASTQ quality assessment, trimming, alignment or transcript-level quantification and gene-level summarisation using FastQC, MultiQC, fastp, STAR, Salmon and featureCounts inside a reproducible Snakemake or Nextflow workflow.

Bulk RNA-seq Statistical Analysis

Domain-specific differential expression for affective symptoms, apathy, agitation and psychosis using DESeq2, edgeR or limma-voom, adjusted for neuropathology, technical covariates and estimated cell composition, with false-discovery-rate control.

Sex & APOE ε4 Interaction Analysis

Test whether molecular associations with each symptom domain differ by biological sex or APOE ε4 carriage — reporting stratified estimates only where sample size and model stability are adequate.

Cell-Composition Deconvolution & Cellular Mapping

Estimate the contribution of excitatory and inhibitory neurons, microglia, astrocytes, oligodendrocytes and endothelial cells to determine whether signals reflect altered abundance, altered regulation, or both.

Network Biology & Single-Nucleus Validation

WGCNA co-expression modules, Gene Ontology and Reactome enrichment, STRING/Cytoscape network mapping, and donor-level pseudobulk validation across SEA-AD, ssREAD and independent GEO single-nucleus datasets.

Explainable Machine Learning & Translational Prioritisation

Nested cross-validation with elastic-net, random forest, SVM and XGBoost, SHAP-based interpretation, a compact 10–30 gene panel, and drug-target mapping through DGIdb, Open Targets, DrugCentral and ChEMBL.

Click any row to expand domain details.

DomainRepresentative ManifestationsKey MetricsScope

AffectiveDepression, anxiety, dysphoria, emotional distressMean / max severity, persistence, cumulative burdenCore

Affective symptoms — depression, anxiety, dysphoria and emotional distress — are modelled as quantitative traits: mean severity, maximum severity, persistence across consecutive assessments and time-weighted cumulative burden, rather than a single yes/no label.

ApathyReduced motivation, reduced initiative, emotional indifferencePersistent apathy, longitudinal slope, high vs low burdenCore

Apathy is captured through persistence across visits, the rate of worsening or improvement, and high-burden versus low-burden classification — allowing molecular programmes to be linked to how apathy actually unfolds over time.

AgitationIrritability, aggression, restlessness, disinhibitionEpisode frequency, severity trajectory, cumulative burdenCore

Agitation — irritability, aggression, restlessness and disinhibition — is characterised by episode frequency, severity trajectory and cumulative burden, so that patients with comparable neuropathology but very different behavioural courses can be compared directly.

PsychosisDelusions, hallucinations, paranoid ideationPresence, persistence, onset timing, severityCore

Psychosis — delusions, hallucinations and paranoid ideation — is modelled through presence, persistence, onset timing and severity. A psychosis-positive versus psychosis-negative contrast is one candidate outcome for the explainable prediction module.

Sleep (optional)Circadian disruption, nocturnal behaviour, sleep disturbanceRepeated sleep-related symptom burdenOptional

Sleep-related disturbance is included as an optional domain, subject to cohort availability and the quality of repeated sleep-symptom measurements.

Deliverable	Description	Value
Curated Clinical–Expression Dataset	Harmonised symptom, demographic, pathology and gene-expression data	✓ Transparent analysis
NGS Analysis Pipeline	Documented QC, quantification and statistical workflow	✓ Portfolio-ready competency
Domain-Specific Signatures	Gene lists for affective symptoms, apathy, agitation and psychosis	✓ Mechanistic differentiation
Shared BPSD Signature	Cross-domain molecular programme	✓ Common therapeutic biology
Sex/APOE-Stratified Findings	Interaction estimates and subgroup-specific results	✓ Precision psychiatry
Cell-Type Map	Neural and glial localisation of candidate signals	✓ Biological interpretability
Co-expression & Regulatory Networks	Modules, hubs and enriched pathways	✓ Systems-level insight
Predictive Gene Panel	Compact and explainable candidate signature	✓ Biomarker hypothesis
Translational Target Map	Prioritised pathways, proteins and drug-target relationships	✓ Future therapeutics
Manuscript Package	Abstract, figures, tables, methods, results framework and references	✓ Publication & conferences
Training Dossier	Code repository, data dictionary, presentation and technical report	✓ Academic & career development

Core Analytical Layer — Included in This Project

Sex- and APOE ε4-Stratified Precision Psychiatry

Biological sex and APOE ε4 genotype can modify both Alzheimer’s disease progression and psychiatric manifestation, yet these interactions are rarely built into symptom-focused transcriptomic models. This module adds interaction models — symptom burden × sex, symptom burden × APOE ε4 status and symptom burden × neuropathological severity — to the core pipeline, with cautious reporting wherever subgroup sample sizes are limited.

Participants completing it learn to convert psychiatric observations into measurable phenotypes, evaluate confounding, and communicate clinically restrained conclusions.

Discovery & Validation Resources

Resource	Role in the Project	Stage
ROSMAP / AMP-AD	Principal discovery resource — longitudinal clinical evaluation, behavioural assessment, postmortem neuropathology, genotype and RNA-seq	✓ Discovery
SEA-AD	Single-nucleus and spatially resolved validation across the Alzheimer’s disease pathological spectrum	✓ Validation
ssREAD	Curated discovery and comparison of Alzheimer’s-related single-cell, single-nucleus and spatial datasets	✓ Validation
GSE174367	Large prefrontal-cortex single-nucleus multi-omic dataset for cell-type and disease-stage validation	✓ Validation
GSE138852	Entorhinal-cortex single-nucleus dataset for independent regional validation	✓ Validation
MIT ROSMAP single-nucleus	Optional matched or related single-nucleus validation within ROSMAP-derived samples, subject to access	✓ Optional

Fees & Payment Plan

Flexible 3-Stage Payment — NGS + Publication Track

The total program fee is split across three instalments so each payment is tied to tangible research progress. All payments are made in Australian Dollars (AUD) through Stripe.

Total program fee

$1,750 (AUD)

approx. $1,200 (USD)

Track

NGS + Publication

6-month research project

Duration

6 months

3 payment stages

Payment method

Stripe

Charged in AUD

Payment distribution

Stage 1

A$590

34% · at registration

Stage 2

A$590

34% · after 2 months

Stage 3

A$570

32% · after 5 months

Month-by-month view

Month 1

Protocol & Data

A$590 due

Month 2

RNA-seq QC

—

Month 3

Association

A$590 due

Month 4

Cells & Networks

—

Month 5

Validation

—

Month 6

Manuscript

A$570 due

Stage 1 window

Stage 2 window

Stage 3 window

Stage-by-stage breakdown — click to expand

Stage 1 — Protocol, Data Access & RNA-seq QC A$590 34% Expand ⌄

Due at registration · Covers months 1–2

Literature ReviewData AccessPhenotype FrameworkBulk RNA-seq QC

Welcome kit — software setup guide, reading list and data-access guidance for ROSMAP / AMP-AD
Month 1 — literature review, data-access applications, neuropsychiatric phenotype framework, metadata dictionary and reproducible environment
Month 2 — bulk RNA-seq quality control, count-matrix preparation, exploratory analysis, sample exclusions, covariate and batch assessment
Milestones — approved protocol with harmonised metadata schema, and an analysis-ready discovery dataset
Stage-gate reviews at the end of Month 1 (data access and feasibility) and Month 2 (covariates and exclusion criteria)
Weekly 1:1 Zoom sessions with assigned mentor, with session recordings provided
Access to a dedicated project support group

Why A$590 at this stage? The first payment secures your seat and activates onboarding — data-access support, protocol design, the metadata schema and the reproducible analysis environment. This groundwork happens before any statistics are run, and it is what makes the later analysis defensible. Stage 1 is non-refundable once onboarding has commenced.

Stage 2 — Association, Interaction & Cellular Validation A$590 34% Expand ⌄

Due after 2 months (start of month 3) · Covers months 3–5

Differential ExpressionSex / APOE InteractionsDeconvolutionSingle-Nucleus Validation

Month 3 — domain-specific differential expression, longitudinal symptom modelling, and sex and APOE ε4 interaction analyses
Month 4 — cell-composition deconvolution, pathway enrichment, co-expression and regulatory-network analysis
Month 5 — single-nucleus RNA-seq processing, donor-level pseudobulk analysis and cell-type-specific validation
Milestones — primary molecular association results, a biological-module and cell-composition map, and an independent cellular validation package
Stage-gate reviews at the end of Month 3 (select credible symptom domains) and Month 5 (confirm which genes and modules pass validation)
Weekly 1:1 Zoom sessions with mentor
Mid-project scientific review and written feedback from guide

Why A$590 at this stage? By month 3 you hold an analysis-ready dataset and an approved protocol — tangible evidence of progress. This payment covers the scientific core of the project: three months of interaction modelling, cell deconvolution, network analysis and single-nucleus validation, where most of the mentor’s review time is spent.

Stage 3 — Prediction, Translation & Manuscript Delivery A$570 32% Expand ⌄

Due after 5 months (start of month 6) · Covers month 6

Explainable MLDruggabilityManuscriptReproducibility Archive

Month 6 — explainable prediction, druggability analysis, and final figures and tables
Manuscript, presentation and repository documentation — a submission-ready research package
Final stage-gate: freeze the analysis, manuscript figures and reproducibility archive
Training dossier — code repository, data dictionary, presentation and technical report
Official BDG Lifesciences certificate of completion
Personalised letter of recommendation from your guide
Weekly 1:1 Zoom sessions with mentor through to manuscript submission

Why A$570 — the final stage — at this point? By the end of month 5 you have validated results and a clear picture of the scientific and professional value built. The final, slightly lower payment carries you through prediction, translational mapping and manuscript preparation, and releases the career-defining deliverables: the manuscript package, certificate and letter of recommendation.

Payment summary

Stage	When to Pay	Amount (AUD)	% of Total	Unlocks
Stage 1	At registration	A$590	34%	Onboarding, protocol, data access, RNA-seq QC
Stage 2	After 2 months	A$590	34%	Association, interaction and single-nucleus validation
Stage 3	After 5 months	A$570	32%	Prediction, manuscript, certificate, recommendation
Total	Over 6 months	A$1,750	100%	Complete NGS + Publication research project

Payment terms: The total fee is A$1,750 (approx. US$1,200) and all payments are made in Australian Dollars (AUD) through Stripe. Stage 1 (A$590) is non-refundable once onboarding has commenced. Stage 2 (A$590) is due after 2 months for participants in good standing. Stage 3 (A$570) is due after 5 months and releases the manuscript package, certificate and letter of recommendation. All fees are non-transferable. Full details are in the Terms & Conditions below. For queries, contact research@bdglifesciences.com

Terms & Conditions

Fees & currency. The total program fee is A$1,750 (Australian Dollars), indicatively equivalent to US$1,200. All payments are charged in Australian Dollars (AUD) and processed through Stripe. The US dollar equivalent is shown for reference only; your bank or card issuer may apply its own exchange rate and fees.
Instalments. Stage 1: A$590 at registration. Stage 2: A$590, due after 2 months of the project (start of month 3). Stage 3: A$570, due after 5 months of the project (start of month 6). Stage 2 is due for participants in good standing.
Secure checkout. Payments are made through Stripe’s secure checkout. Card details are handled by Stripe and are not stored by BDG Lifesciences.
Refunds & transfers. Stage 1 is non-refundable once onboarding has commenced, as it covers data-access support, protocol design, software setup and mentor allocation. All fees are non-transferable.
Stage 3 deliverables. Manuscript finalisation, certificate issuance and the letter of recommendation are released upon confirmation of Stage 3 payment.
Data access. Some datasets (e.g. ROSMAP symptom-level variables) are controlled-access and depend on approval by the data custodians. Participants must comply with all data-use agreements, must not attempt re-identification, and must not export restricted individual-level information.
Research outcomes. BDG Lifesciences supports manuscript preparation, but no journal acceptance can be guaranteed. Findings are research findings, not clinical diagnostic tests.
Queries. Contact research@bdglifesciences.com before making payment if you have any questions about these terms.

Common Questions

Frequently Asked Questions

Rather than a conventional Alzheimer’s-versus-control comparison, the project asks why patients with comparable neurodegenerative pathology develop different psychiatric and behavioural manifestations — depression, apathy, agitation, psychosis and, optionally, sleep disturbance — and which brain-cell-specific transcriptional programmes are associated with them, including how sex and APOE ε4 status modify these programmes.

It is a real, retrospective, computational research project. You work with clinically phenotyped bulk RNA-seq and independent single-nucleus datasets, follow a pre-specified analysis plan with stage-gate reviews, and finish with a submission-ready research package including a manuscript framework, figures and a reproducible code repository.

The principal discovery resource is ROSMAP (via the AMP-AD / AD Knowledge Portal), with SEA-AD, ssREAD and GEO datasets GSE174367 and GSE138852 used for single-nucleus validation. ROSMAP symptom-level variables are controlled-access, so data-access applications are submitted in Month 1 and a secondary public-dataset plan is maintained in case of delay.

The project is designed for a medicine-oriented student pursuing future training in psychiatry, and that clinical framing runs through it. Life science, neuroscience and bioinformatics graduates and professionals are equally welcome; a background in medicine or life sciences is helpful, and the analytical tools and methods are introduced as part of the project.

FastQC, MultiQC, fastp, STAR, Salmon, featureCounts, DESeq2 / edgeR / limma-voom, sva or RUVSeq, WGCNA, STRING and Cytoscape, plus R and Python with Snakemake or Nextflow for reproducible workflows. Machine-learning and translational modules use elastic-net, random forest, SVM, XGBoost, SHAP, DGIdb, Open Targets, DrugCentral and ChEMBL.

The project is built around a defensible publication pathway: longitudinal symptom phenotyping, bulk RNA-seq association analysis, sex/APOE effect modification and independent cell-type validation. No journal acceptance can be guaranteed — publication strength depends on access to adequately phenotyped data and rigorous independent validation.

The total fee is A$1,750 (approximately US$1,200), paid in three instalments — A$590 at registration, A$590 after 2 months and A$570 after 5 months. All payments are made in Australian Dollars (AUD) through Stripe. The US dollar figure is indicative only; see the Terms & Conditions in the payment section.

Payments are processed in AUD on Stripe. If your card or bank account is in another currency, your card issuer converts the amount at its own rate and may charge conversion fees, which are outside BDG Lifesciences’ control.

It uses retrospective postmortem data, so results are research findings rather than clinical diagnostic tests, and computational drug-target mapping does not imply therapeutic efficacy. Sex- and APOE-stratified estimates are reported only where sample sizes allow, and the project plan documents each limitation with a mitigation strategy.

Unlike generic internship programmes, Project #95 is built around a defined, unsolved research question with a stage-gated plan and a publication outcome. With 88+ internationally published projects and 15+ years of experience, BDG Lifesciences delivers scientific credibility backed by a proven track record.

Team for Current / Ongoing Research Projects

94.

In Silico Prioritization of TEAD-Targeted Degrader-Inspired Molecules for Malignant Pleural Mesothelioma Using Molecular Docking, ADMET Screening, Molecular Dynamics Simulation, and Binding Free-Energy Analysis

Ojasi Kawadkar, Fremont, CA, USA.

Sharvari Kulkarni, Chief Technical Officer, BDG Lifesciences, India.

Syed Asif Husain, BBR Pty Ltd & BDG Lifesciences, India.

93.

Integrative Single-Cell Transcriptomics and CRISPR-Cas9 Perturbation Modeling for Regulatory Network Mapping in the Tumour Microenvironment

Isabella Montemayor, Keystone School, San Antonio, USA.

Ayanjeet Chowdhury, BDG Lifesciences, India.

Syed Asif Husain, BBR Pty Ltd & BDG Lifesciences, India.

92.

Mapping Druggable RNA Landscapes: An End-to-End Pipeline from Transcriptomics to Therapeutic Targeting

Yana Joshi, Henry M. Jackson High School, WA, USA.

Sharvari Kulkarni, Chief Technical Officer, BDG Lifesciences, India.

Ayanjeet Chowdhury, Technical Officer, BDG Lifesciences, India.

91.

ImmunoSpatial Blueprint: Mapping Tumor-Immune Interactions in Triple-Negative Breast Cancer Using Spatial Transcriptomics

Dr Malu Sreekumar, Jawaharlal Institute of Post-graduate Medical Education and Research (JIPMER), Puducherry, India.

Tanmay Bandbe, Guru Nanak Khalsa College of Arts, Science & Commerce, Mumbai, India.

90.

Single-cell RNA-seq Analysis of Chemotherapy-Induced Transcriptomic Changes in Pancreatic Ductal Adenocarcinoma (PDAC)

Dr Ilham Mahgoub, Qatar University, Qatar.

Ritam Baruah, Ajanta Pharma Limited, Assam, India.

Aakash Deva Thirukonda Prakash, KTH Royal Institute of Technology, India.

88.

Computational Insights into Selective ERβ Agonists for Benign Prostatic Hyperplasia (BPH) Treatment | A Virtual Screening, Molecular Docking & Simulations Study

Shivani Tuli, University of Pittsburg, Pittsburgh, USA.

Rakshitha Prakash, Vellore Institute of Technology (VIT), India.

Chandrima Das Sinha, Techno India University, India.

Swapnaja Gulawani, Agharkar Research Institute, India.

Trần Thị Thư, Hồ Chí Minh, Vietnam.

87.

Decoding Pan-Cancer Pathogenesis: A Multi-Layered Analysis of Prognostic mRNAs, miRNAs, lncRNAs via Co-Expression Networks and PPINs

Ritam Baruah, Ajanta Pharma Limited, Assam, India.

Tanmay Bandbe, Guru Nanak Khalsa College of Arts, Science & Commerce, Mumbai, India.

86.

Unveiling Autoimmune Genes and Regulatory Elements in Head and Neck Squamous Cell Carcinoma through Advanced Machine Learning and Network-Based Analysis

Mathews Kennedy, Trivandrum, Kerala, India.

Angela Susan Anil, School of Biotechnology, Amrita Vishwa Vidyapeetham, Kerala.

Elham O Mahgoub, Alneelain University, Sudan.

85.

Molecular Modeling study of derivatives of Leaf Extracts of medicinal plant Solanum torvum and Serine/Threonine Kinase from Mycobacterium Tuberculosis

Tenzin Kalsang, Manipal Academy of Higher Education, India.

Elham O Mahgoub, Alneelain University, Sudan.

Laxmi C M, Teesside University, Middlesbrough, UK.

Ilham Said Salman, Beirut Arab University, Lebanon.

84.

Targeting Tumor Progression: Identifying Differentially Expressed Genes and Pathways in Pancreatic Ductal Adenocarcinoma using RNAseq

Kasturi Mukherjee, Kyvor Genomics Pvt. Ltd, Chennai, India.

83.

Next Generation Sequencing | Unraveling the Cancer Code: Gene Expression Profiling with RNAseq

Ilham Omer Mahgoub, Alneelain University, Sudan.

Arnav Kolluru, Liberty High School, Renton, WA, United States.

82.

NGS Data Analysis | Prediction of Multiple Myeloma Using RNASeq Data

Dolly Jagwani, Indian Institute of Science Education and Research, Pune (IISER-P), Maharashtra, India.

Mathews Kennedy, Cancer Biology and Therapeutics, Middlesex University, London, UK.

81.

NGS Data Analysis of Cancer Tissues | A Cancer Biology Study

Pooja Prakash Mankar, Shri Shivaji College of Arts, Commerce and Science, Akola, India.

Amit Kumar, ICMR, New Delhi, India.

Ilham Omer Mahgoub, Alneelain University, Sudan.

Trajectory- and Cell-Type–Resolved Transcriptomic Profiling of Neuropsychiatric Symptoms in Alzheimer’s Disease

Secure Your Place in Project #95