Genes
Gene Notes
This page records genes that repeatedly appear while reading papers. The goal is not to build a complete gene encyclopedia, but to keep practical reading notes: what the gene generally marks, and what it means in the paper where it appears.
Index
| Gene | Full name | Common reading context | First note |
|---|---|---|---|
| PTPRC / CD45 | Protein tyrosine phosphatase receptor type C | Pan-leukocyte marker; immune-cell enrichment or depletion | Patel et al. 2014: CD45+ depletion before GBM single-cell profiling |
| IDH1 | Isocitrate dehydrogenase 1 | Glioma molecular classification; IDH-mutant vs IDH-wildtype | Patel et al. 2014: all sampled primary GBMs were IDH1/2-wildtype |
| IDH2 | Isocitrate dehydrogenase 2 | Glioma molecular classification; IDH-mutant vs IDH-wildtype | Patel et al. 2014: all sampled primary GBMs were IDH1/2-wildtype |
| GZMB | Granzyme B | Cytotoxic lymphocyte effector activity | Zhang et al. 2026: CD8+ T cells in tumour core |
| GZMK | Granzyme K | CD8+ T cell state, often less terminally cytotoxic or peripheral/memory-like context | Zhang et al. 2026: CD8+ T cells in tumour periphery |
| BRAF | B-Raf proto-oncogene, serine/threonine kinase | MAPK pathway oncogene; melanoma targeted therapy | Tirosh et al. 2016: 4/17 genotyped melanomas had activating BRAF mutations |
| NRAS | NRAS proto-oncogene, GTPase | MAPK pathway oncogene; melanoma molecular subtype | Tirosh et al. 2016: 5/17 genotyped melanomas had activating NRAS mutations |
| NF1 | Neurofibromin 1 | RAS pathway negative regulator; melanoma without BRAF/NRAS mutation | Tirosh et al. 2016: NF1-inactivated melanomas are mentioned among BRAF-wildtype therapeutic contexts |
| RAF1 / RAF | Raf-1 proto-oncogene, serine/threonine kinase | MAPK pathway signaling; RAF/MEK inhibition context | Tirosh et al. 2016: RAF and MEK inhibition selects or enriches AXL-high drug-resistant melanoma states |
| MITF | Melanocyte inducing transcription factor | Melanoma lineage program; melanocytic differentiation and proliferative state | Tirosh et al. 2016: MITF-high malignant cells mark one melanoma transcriptional state |
| AXL | AXL receptor tyrosine kinase | Invasive, drug-resistant melanoma state; RTK signaling | Tirosh et al. 2016: AXL-high malignant cells pre-exist treatment and are enriched by RAF/MEK inhibition |
| MKI67 / Ki-67 | Marker of proliferation Ki-67 | Proliferating cells; cell-cycle activity by staining or expression | Tirosh et al. 2016: Ki67 staining validates low- and high-cycling melanoma tumors |
| KDM5B / JARID1B | Lysine demethylase 5B | Slow-cycling or noncycling melanoma cells; drug-tolerant state context | Tirosh et al. 2016: KDM5B correlates with noncycling malignant cells and is mutually exclusive with Ki67 |
| CD14 | CD14 molecule | Monocyte/macrophage lineage marker; classical monocytes | Cheng et al. 2021: CD14 helps distinguish monocyte/macrophage subsets in pan-cancer TIMs |
| FCGR3A / CD16 | Fc gamma receptor IIIa | Non-classical monocytes, NK cells, Fc receptor biology | Cheng et al. 2021: CD16 helps separate monocyte subsets, especially CD14+CD16hi non-classical monocytes |
| TNF | Tumor necrosis factor | Inflammatory cytokine; mast-cell anti-tumor functional axis in TME | Cheng et al. 2021: TNF+ mast cells are used as a proxy for anti-tumor mast-cell tendency |
| VEGFA | Vascular endothelial growth factor A | Angiogenesis; pro-tumor/pro-vascular TME signal | Cheng et al. 2021: VEGFA+ mast cells are used as a proxy for pro-angiogenic mast-cell tendency |
PTPRC / CD45
Full name: Protein tyrosine phosphatase receptor type C
Basic meaning: PTPRC encodes CD45, a receptor-type protein tyrosine phosphatase broadly expressed by leukocytes. In practical single-cell reading, CD45 is usually treated as a pan-leukocyte marker rather than a marker for one specific immune subtype.
How to read it: CD45 positivity usually means the cell belongs to the immune compartment, including T cells, B cells, NK cells, monocytes, macrophages and related leukocyte populations. CD45 negativity is often used experimentally to enrich non-immune compartments, but it does not by itself prove that remaining cells are malignant.
In Patel et al. 2014: The authors depleted CD45+ cells before sorting GBM single cells. The point was to focus the experiment on glioblastoma tumour cells and avoid confusing malignant-cell transcriptional heterogeneity with differences between tumour cells and infiltrating immune cells. Their bulk control was also generated from the CD45-depleted tumour cell population.
Reading caution: CD45 is a broad compartment marker. It is useful for separating immune from non-immune cells, but it does not distinguish T cells from myeloid cells, nor does CD45 depletion remove all possible non-malignant cells. Malignancy still needs additional evidence such as inferred CNV, tumour marker programs or genotype.
IDH1
Full name: Isocitrate dehydrogenase 1
Basic meaning: IDH1 encodes a metabolic enzyme, but in glioma reading its main practical meaning is as a molecular classification marker. IDH1 mutation status helps separate IDH-mutant gliomas from IDH-wildtype glioblastoma.
How to read it: In glioma papers, IDH-mutant usually means a tumor carries a pathogenic IDH1 or IDH2 mutation, most commonly IDH1 R132H. IDH-wildtype means no such IDH mutation is detected. This distinction is not cosmetic: it marks biologically and clinically different glioma entities.
In Patel et al. 2014: The authors state that all five primary glioblastomas were IDH1/2-wildtype. This matters because they are studying intratumoral heterogeneity within classic primary GBM rather than mixing IDH-mutant and IDH-wildtype gliomas.
Reading caution: IDH1 is not a general tumor marker. Its interpretation is highly disease-specific. In glioma, IDH mutation status affects classification, prognosis and epigenetic state, so it should be checked before comparing tumors.
IDH2
Full name: Isocitrate dehydrogenase 2
Basic meaning: IDH2 is the mitochondrial paralog of IDH1. In glioma papers it is usually read together with IDH1 as part of IDH1/2 mutation status.
How to read it: IDH2 mutations are less common than IDH1 mutations in glioma, but they serve the same broad classification purpose: distinguishing IDH-mutant gliomas from IDH-wildtype glioblastoma.
In Patel et al. 2014: IDH1/2-wildtype means neither IDH1 nor IDH2 mutation was detected in the sampled primary GBMs. The authors use this to keep the cohort within the IDH-wildtype primary glioblastoma context.
Reading caution: When a paper says IDH-wildtype, it usually implies both IDH1 and IDH2 were considered. Do not interpret IDH2 alone without checking the paired IDH1 status and the glioma classification context.
GZMB
Full name: Granzyme B
Basic meaning: GZMB encodes granzyme B, a cytotoxic serine protease used by cytotoxic lymphocytes, especially CD8+ T cells and NK cells, to kill target cells.
How to read it: In tumour immunology papers, high GZMB often points to cytotoxic effector activity. It is commonly read together with genes such as PRF1, NKG7 and GNLY.
In Zhang et al. 2026: The paper uses GZMB as an example of a spatially polarized CD8+ T cell program. CD8+ T cells preferentially express GZMB when localized to the tumour core, suggesting that spatial context is associated with a more cytotoxic effector-like state.
Reading caution: GZMB is a useful cytotoxic marker, but one gene alone does not prove effective tumour killing. It should be interpreted with broader T cell state markers, exhaustion markers, antigen presentation context and spatial location.
GZMK
Full name: Granzyme K
Basic meaning: GZMK encodes granzyme K, another granzyme expressed by subsets of T cells, especially certain CD8+ T cell states.
How to read it: In single-cell and tumour immunology papers, GZMK often appears in CD8+ T cell states that are not the same as strongly GZMB-high terminal cytotoxic states. Depending on context, it may mark memory-like, transitional, inflammatory or peripheral CD8+ T cell programs.
In Zhang et al. 2026: The paper contrasts GZMK with GZMB. CD8+ T cells preferentially express GZMK when localized to the tumour periphery, supporting the idea that CD8+ T cell state varies with spatial location.
Reading caution: GZMK should not be simplified as merely “weaker GZMB.” Its meaning depends on tissue, disease, activation state and the surrounding gene program.
BRAF
Full name: B-Raf proto-oncogene, serine/threonine kinase
Basic meaning: BRAF encodes a RAF-family kinase in the MAPK signaling pathway. In melanoma reading, BRAF is most often encountered through activating mutations such as BRAF V600E, which can drive MAPK pathway activity and create a targeted-therapy context.
How to read it: BRAF-mutant melanoma usually means the tumor carries an activating BRAF alteration and may be treated with RAF and MEK inhibitors. BRAF-wildtype does not mean MAPK signaling is irrelevant; signaling may instead be driven by NRAS mutation, NF1 loss, RAF fusions or other alterations.
In Tirosh et al. 2016: Genotypic information was available for 17 of 19 melanomas, and 4 had activating BRAF mutations. The paper discusses BRAF V600E melanomas because RAF/MEK inhibitor therapy can produce responses but resistance commonly emerges. Their single-cell analysis highlights AXL-high malignant-cell states that may pre-exist treatment and become enriched after MAPK-targeted therapy.
Reading caution: BRAF status is not the same as cell state. A BRAF-mutant tumor can contain MITF-high and AXL-high malignant cells, and a bulk genotype does not reveal which single-cell subpopulations are poised for drug resistance.
NRAS
Full name: NRAS proto-oncogene, GTPase
Basic meaning: NRAS encodes a small GTPase upstream of RAF/MEK/ERK signaling. Activating NRAS mutations are common melanoma drivers and are usually read as part of MAPK pathway activation.
How to read it: In melanoma, NRAS mutation often marks tumors that are not BRAF-mutant but still use MAPK signaling. Unlike BRAF V600E melanoma, NRAS-mutant melanoma has historically been harder to target directly, so papers often discuss it as part of the broader need for better therapeutic strategies.
In Tirosh et al. 2016: Of the 17 genotyped melanomas, 5 had activating NRAS mutations. The cohort therefore includes BRAF-mutant, NRAS-mutant and BRAF/NRAS-wildtype tumors, allowing the authors to frame single-cell state heterogeneity across multiple melanoma genetic backgrounds.
Reading caution: NRAS mutation status is a driver-genotype label, not a complete description of transcriptional phenotype. Single-cell MITF/AXL state, cell cycle status and microenvironment composition can vary within and across NRAS-mutant tumors.
NF1
Full name: Neurofibromin 1
Basic meaning: NF1 encodes neurofibromin, a negative regulator of RAS signaling. Loss or inactivation of NF1 can increase RAS/MAPK pathway activity.
How to read it: In melanoma papers, NF1 often appears in the context of BRAF/NRAS-wildtype tumors or as another route into MAPK pathway activation. It helps explain why tumors without BRAF or NRAS activating mutations may still have MAPK signaling dependence or altered therapeutic vulnerabilities.
In Tirosh et al. 2016: NF1 is mentioned when the authors describe melanoma patients without BRAF mutations, including tumors with NRAS mutations, NF1 inactivation or rarer events such as RAF fusions. The point is therapeutic: patients outside the BRAF-mutant category lack the same established targeted-therapy options.
Reading caution: NF1 is a tumor suppressor context, not a marker of one specific cell type. Interpret NF1 together with broader genotype, MAPK pathway activity and the malignant-cell transcriptional state.
RAF1 / RAF
Full name: Raf-1 proto-oncogene, serine/threonine kinase
Basic meaning: RAF1, also called CRAF, is a RAF-family kinase in the MAPK pathway. In cancer papers, RAF may refer generically to the RAF kinase family or to RAF inhibitor treatment rather than to the RAF1 gene specifically.
How to read it: When a paper says RAF and MEK inhibition, it usually means pharmacologic blockade of the MAPK pathway, commonly in BRAF-mutant melanoma. When it says RAF fusion, it refers to structural alterations involving RAF-family kinases that can activate signaling.
In Tirosh et al. 2016: The authors discuss RAF and MEK inhibitors as standard targeted therapy for BRAF V600E melanoma. They show that melanoma tumors and cell lines can contain AXL-high drug-resistant subpopulations before treatment and that these populations become enriched after RAF/MEK inhibition.
Reading caution: RAF is sometimes a pathway/drug-class shorthand rather than a precise gene symbol. If the paper needs gene-level interpretation, check whether it means BRAF, RAF1/CRAF, ARAF, a RAF fusion or a RAF inhibitor.
MITF
Full name: Melanocyte inducing transcription factor
Basic meaning: MITF is a lineage-associated transcription factor in melanocytes and melanoma. In melanoma papers, it often marks a melanocytic differentiation program and is linked to a more proliferative, lineage-identity-high malignant-cell state.
How to read it: MITF-high melanoma cells are usually interpreted as retaining melanocytic lineage identity. They often express melanocyte differentiation genes and can contrast with AXL-high, invasive or therapy-resistant states.
In Tirosh et al. 2016: MITF is used as one pole of melanoma malignant-cell heterogeneity. The paper highlights that individual tumors can contain both MITF-high and AXL-high malignant cells, showing that melanoma transcriptional states vary within the same tumor rather than only across patients.
Reading caution: MITF is not just a marker gene; it is a transcription factor tied to melanoma lineage state. A MITF-high signal should be interpreted with the broader melanoma program, proliferation state and treatment context.
AXL
Full name: AXL receptor tyrosine kinase
Basic meaning: AXL encodes a receptor tyrosine kinase. In melanoma reading, AXL often marks an invasive, mesenchymal-like or drug-resistant malignant-cell state, especially in contrast to MITF-high melanoma states.
How to read it: AXL-high melanoma cells are commonly associated with reduced melanocytic differentiation, increased invasion-related programs and resistance to MAPK pathway inhibition. In single-cell data, AXL is important because it can identify a minority malignant-cell state that may be missed or averaged out in bulk data.
In Tirosh et al. 2016: The authors show that AXL-high malignant cells can pre-exist RAF/MEK inhibitor treatment and become enriched after treatment. This connects intratumoral transcriptional heterogeneity to therapy resistance: resistant states do not necessarily have to be newly created by therapy; they may already exist as rare subpopulations.
Reading caution: AXL-high should not be reduced to one marker alone. It is most meaningful when supported by a broader transcriptional program and by its contrast with MITF-high melanoma lineage states.
MKI67 / Ki-67
Full name: Marker of proliferation Ki-67
Basic meaning: MKI67 encodes Ki-67, a nuclear protein widely used as a marker of cell proliferation. In pathology and single-cell reading, Ki-67 positivity usually indicates cells in active cell cycle rather than quiescent or noncycling cells.
How to read it: Ki-67 is often used by immunostaining to estimate the proliferative fraction of a tumor. In single-cell RNA-seq, MKI67 may appear with broader G1/S and G2/M gene signatures, but it is usually safer to interpret proliferation from a full cell-cycle program rather than from one gene alone.
In Tirosh et al. 2016: The authors used cell-cycle signatures to infer cycling malignant melanoma cells and then compared low-cycling and high-cycling tumors with Ki67+ staining. Ki67 staining supported the single-cell estimate that tumors such as Mel79 had low cycling fractions, whereas tumors such as Mel78 had higher cycling fractions.
Reading caution: Ki-67 marks proliferation, not malignancy. Immune and stromal cells can also proliferate, so in tumor single-cell data it should be interpreted after cell type or malignant-cell classification.
KDM5B / JARID1B
Full name: Lysine demethylase 5B
Basic meaning: KDM5B, also known as JARID1B, encodes a histone demethylase. In melanoma literature, it is often associated with slow-cycling, noncycling or drug-tolerant malignant-cell states.
How to read it: KDM5B-high melanoma cells are commonly interpreted in the context of phenotypic heterogeneity and therapy tolerance. It does not simply mark absence of proliferation; it points to a chromatin-associated state that may help cells persist under stress or treatment.
In Tirosh et al. 2016: KDM5B was correlated with noncycling malignant cells. The authors also note that KDM5B/JARID1B and Ki67 were mutually exclusive by immunofluorescence, supporting the idea that some melanoma cells occupy a slow-cycling or noncycling state.
Reading caution: KDM5B should be read as part of a broader cell-state program. A single KDM5B-high signal is not enough to prove a stable cancer stem cell or resistant clone without supporting transcriptional, functional or treatment-response evidence.
CD14
Full name: CD14 molecule
Basic meaning: CD14 encodes a co-receptor involved in innate immune recognition, especially in monocytes and macrophages. In single-cell reading, CD14 is most often used as a monocyte/macrophage lineage marker rather than as a disease-specific gene.
How to read it: High CD14 commonly marks classical monocytes and monocyte/macrophage-like myeloid cells. It is often interpreted together with LYZ, FCN1, S100A8, S100A9, CD68, CD163, HLA genes and FCGR3A/CD16 to distinguish blood monocytes, tissue-infiltrating monocytes and macrophage states.
In Cheng et al. 2021: The authors use CD14 and CD16 expression to classify monocyte subsets in the pan-cancer tumor-infiltrating myeloid atlas. Classical monocytes are described as CD14 high CD16 negative, while non-classical monocytes are CD14 positive CD16 high. Tissue-derived monocytes show additional macrophage- and antigen-presentation-related programs.
Reading caution: CD14 is not specific to one fixed cell state. In tumors, CD14 expression can appear across monocyte-to-macrophage continua, so it should be interpreted with broader myeloid markers, tissue source and functional genes.
FCGR3A / CD16
Full name: Fc gamma receptor IIIa
Basic meaning: FCGR3A encodes CD16a, an Fc receptor expressed by non-classical monocytes, NK cells and some macrophage-related populations. In immunology papers, CD16 can refer to Fc gamma receptor III, and gene-level interpretation usually maps to FCGR3A in human single-cell RNA-seq.
How to read it: In monocyte annotation, FCGR3A/CD16 helps distinguish non-classical monocytes from classical CD14-high monocytes. In NK-cell contexts, CD16 is also important for antibody-dependent cellular cytotoxicity, so the surrounding markers determine whether it is being used for monocytes, NK cells or other Fc-receptor-positive cells.
In Cheng et al. 2021: CD16 is used with CD14 to define monocyte subsets across tumor-infiltrating myeloid cells. The paper describes classical CD14 high CD16 negative monocytes, non-classical CD14 positive CD16 high monocytes, and intermediate CD14 high CD16 positive monocytes in CRC and myeloma where blood-derived monocytes are abundant.
Reading caution: CD16 is not a single-cell-type marker. FCGR3A can mark non-classical monocytes, NK cells or Fc-receptor-expressing myeloid states, so it must be read with CD14, LYZ, NK markers such as NKG7/GNLY, and the broader cluster context.
TNF
Full name: Tumor necrosis factor
Basic meaning: TNF encodes a major inflammatory cytokine produced by multiple immune cell types, including macrophages, T cells and mast cells. It can promote inflammation, endothelial activation, immune-cell recruitment and context-dependent cell death or survival responses.
How to read it: TNF is not a lineage marker. It is better read as an inflammatory functional gene whose meaning depends on the producing cell type and tissue context. In some settings it supports anti-microbial or anti-tumor immunity; in others, chronic TNF signaling may contribute to inflammation-associated pathology.
In Cheng et al. 2021: The authors use TNF+ mast cells as one side of a simplified mast-cell functional axis. Based on prior work that mast-cell-derived TNF can support DC function and T cell priming, they interpret a higher TNF+ / VEGFA+ mast-cell ratio as a proxy for a more anti-tumor mast-cell tendency, especially in NPC.
Reading caution: TNF+ does not by itself prove anti-tumor function. It should be interpreted with the cell type, local immune context, DC/T cell activity and clinical association. In Cheng et al. 2021, it is a useful functional proxy, not a complete mechanism.
VEGFA
Full name: Vascular endothelial growth factor A
Basic meaning: VEGFA encodes a major pro-angiogenic factor that promotes blood vessel growth and vascular permeability. In tumor biology, VEGFA is often read as part of angiogenesis, hypoxia response and pro-tumor vascular remodeling.
How to read it: VEGFA can be expressed by tumor cells, stromal cells and immune cells. Its meaning depends on the source. In TME reading, VEGFA-high signals often suggest angiogenic or vascular-remodeling programs, but the producing cell type should always be checked.
In Cheng et al. 2021: The authors use VEGFA+ mast cells as the pro-angiogenic side of the mast-cell functional axis. Most cancer types had more VEGFA+ than TNF+ mast cells, suggesting that tumor mast cells often have a pro-angiogenic or pro-tumor tendency. NPC was an exception, with a higher TNF+ / VEGFA+ ratio.
Reading caution: VEGFA+ does not prove that a mast cell is functionally driving angiogenesis. It is a marker-based proxy that should be supported by broader angiogenesis programs, spatial proximity to vessels and functional validation.