Services for Biotech and Pharma R&D
Insilico Biosystems provides consulting services for biotech and pharma R&D at all stages of drug discovery and data science tocology development. We are experts in primary, secondary and tertiary analysis of transcriptomics (RNA-Seq, single cell and spatial), proteomics, metabolomics, genetics, electronic medical records, literature and other data types for individual and across projects, . We have broad experience in engineering, curating, integrating and mining data from internal and external sources and building databases and tools.
Please click on each service to see case examples
Compound Profiling & Safety Assessment
1. Profiling small molecular compounds in animal models, by single cell RNA-Seq
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
Cell Metablism. 2022 Jul 5;34(7):1064-1078. “Mapping the Single Cell Transcriptomic Response of Murine Diabetic Kidney Disease to Therapies”
Compound Profiling & Safety Assessment
2. Profiling antibody compound in animal models, by bulk RNA-Seq
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
PLoS One. 2019 Nov 11;14(11):e0223918. “The CSF-1-receptor inhibitor, JNJ-40346527 (PRV-6527), reduced inflammatory macrophage recruitment to the intestinal mucosa and suppressed murine T cell mediated colitis”
Compound Profiling & Safety Assessment
3. Safety assessment of a antibody compound in clinical development
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
N Engl J Med. 2022 Dec 5;387(24):2232-2244. “Talquetamab, a T Cell Redirecting GPRC5D Bispecific Antibody for Multiple Myeloma”
Compound Profiling & Safety Assessment
4. Safety assessment of a cell therapy after BLA approval
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
J Histochem Cytochem, 2022 Apr;70(4):273-287. “Comprehensive BCMA expression profiling in adult normal human brain suggests a low risk of on-target neurotoxicity in BCMA-targeting multiple myeloma therapy”
Target/Biomarker ID & Validation
1. Target Identification through Systematical Mining of Integrated Data (Oncology)
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
Omicsoft User Group Meeting, 2017, Cambridge, MA. “Mining tumor-specific cell surface protein genes and alterations”. TCGA’s paired normal tissues are not truly normal and the coverage is limited, so integration with large normal tissue data collection such as GTEx ensures identifying large number of truly tumor-specific genes and alterations.
Integrating large number of disease studies from internal and external sources provides high statistic power to detect true disease signals among various confounding factors and noises.
Target/Biomarker ID & Validation
Front Genet. 2019 April 30;10:396. “Integrative analysis of DiseaseLand Omics Database for Disease Signatures and Treatments: A Bipolar Case Study”
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
2. Target Identification through Systematical Mining of Integrated Data (Neurosciences)
Integrating large number of disease studies from internal and external sources provides higher statistic power to detect true disease signals among various confounding noises.
Target/Biomarker ID & Validation
3. Target Validation through In-Silico and Wet Lab Assessment.
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
1. In-Silico Validation
2. POC Study in Model System
ACNP 58th Annual Meeting, 2019, Orlando, FL. ”Meta-Analysis Identification and Modeling of a Bipolar Disorder Risk Mutation in LRP8 Using Human Induced Pluripotent Stem Cells"
In-Silico evaluation of candidate targets by utilizing genetic (GWAS, PheWAS, eQTL, …), pathway, literature and other data, followed by proof-of-concept studies in model systems.
Target/Off-Target Safety Assessment
Target safety assessment is part of target identification and validation at early stage. Compound specific safety issues, such as on-target/off-tumor and off-target, may arise as a project advances. A compound of any modality often binds to off-targets in addition to its primary target(s). Different approaches shall be applied to identify off-targets and to address safety consequences of such off-target perturbation along portfolio stages.
Br J Pharmacol. 2023 Feb;180(4):401-421. “Identification and characterization of select oxysterols as ligands for GPR17”. GPR17 is a target for psychiatric disease indication. Bulk and single cell RNA-Seq of comprehensive human and mouse tissues identified GPR17 is specifically expressed in oligodendrocyte and its progenitor cells in brain among all tissue and cell types. It determined that the target will unlikely have mechanistic safety issue and does not need intracerebral drug delivery.
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
1. GPR17 is specifically expressed in brain tissues
A. GTEx (V8) RNA-Seq data released by Broad Institute shows that GPR17 is broadly expressed.(https://gtexportal.org/home/gene/GPR17)
B. Insilico Biosystems reprocessed the same raw data set using new QC and expression quantification methods. It identified that GPR17 is specifically expressed in brain
2. GPR17 is specifically expressed in oligodendrocyte progenitor cells and higher in infant than adult
Disease Profiling & Subtyping
1. Profiling key disease genes using transcriptomics and literature data
Elife. 2020 May 28;9:e58040. “Knowledge synthesis of 100 million biomedical documents augments the deep expression profiling of coronavirus receptors”
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
Disease Profiling & Subtyping
2. Subtyping disease based on epigenetic modification
Diabetologia. 2016 Nov;59(11):2393-2405. “The epigenetic signature of systemic insulin resistance in obese women”
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
Disease Profiling & Subtyping
3. Mimicking disease signaling in model systems of Neuroinflammation
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
Sci Rep. 2021 May 17;11(1):1-14. “Mouse primary microglia respond differently to LPS and poly (I:C) in vitro”
Model Selection & Assay Development
Model systems, such as cell (cell line, Primary and iPSC), 3D spheroid, organoid, organ-on-chip, CDX/PDX, animal (from rodent to NHP), are commonly used in target validation and drug discovery. But their translatability to human often remain unclear. Systematic assessment of a model’s similarity to human (in overall disease profiling, specific disease pathway, gene network of a drug target, et.al.) is essential to select most suitable model system and improve translation.
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
2. Assessing model systems and assay methods by disease marker genes
NIH/NIMH National Cooperative Reprogrammed Cell Research Groups (NCRCRG) to Study Mental Illness (U19), 2019. Disrupted-in-Schizophrenia 1 (DISC1) gene mutation is a strong risk factor for schizophrenia. Is DICS1 mutant iPSC a good cellular model for studying schizophrenia disease? In this project, DISC1 mutant iPSC lines were created from patients with the mutation and isogenic wild-type lines through CRISPR repair. iPSC lines were differentiated to GABAergic-like and Glutamatergic-like neurons. RNA-Seq assay were then applied to identify differentially expressed genes (DEGs) between mutant and wild-type neutrons. Finally, DEGs were compared to thousands of studies using human disease samples for assessing enrichment with disease genes in each study.
1. Validating model systems by full-genome comparison
J Neuroinflammation. 2018 May 22;15(1):153. “RNA sequencing analysis reveals quiescent microglia isolation methods from postnatal mouse brains and limitations of BV2 cells”
3. Species selection by target sequence and expression analysis
Omicsoft User Group Meeting, 2017, Cambridge, MA. Multiple pharma companies worked on Sodium-Glucose Transporter 1 (SGLT1) as a target for diabetes indications. The protein is highly conserved across species (Mouse: 87.8%, Rat: 87.7%, Dog: 86.5%, Cyno: 95.9% and Chimpanzee: 98.8% when compared to human). But in cardiovascular tissues, it’s only expressed in Hominidae and not event in Rhesus or Cynomolgus monkey. Therefore, cardiotoxicity studies using common nonclinical species do not address on-target safety questions.
Big Data Management & Computing
Big data is big in volume at early processing stages. Its final findings in executive summary can be a one-pager. Along many stages of data flow from big raw data to one-pager executive summary, there are different requirements on storage, computing, user access, compliance, and others.
Bio-IT World Conference & Expo, 2014. “Multi-tier Big Data Computing and Management”
Examples illustrated in our services are from published work authored by Insilico Biosystems team members
