METTL3 uncouples chromatin accessibility from transcription during retinal development

Probably not a substantial claim: may not pass S3 (load-bearing).

Similar findings were reported for Mettl14 inactivation and triple deletion of m6A readers Ythdf1, Ythdf2, and Ythdf3.

“Similar findings were reported for Mettl14 inactivation and triple deletion of m6A readers Ythdf1, Ythdf2, and Ythdf3 (Niu et al., 2022; Li et al., 2023).”

• cites Niu et al., 2022 sourceSupported
• cites Li et al., 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

METTL3 and METTL14 associate in an m6A methyltransferase complex to modify RNA, which regulates RNA metabolism, including stability.

“RNA methyltransferases (“writers”) such as METTL3 and METTL14 associate in an N6-methyladenosine (m6A) methyltransferase complex to modify RNA, which regulates RNA metabolism, including stability (Liu et al., 2014).”

• cites Liu et al., 2014 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Overexpression of Zfp292 phenocopied the effects of Mettl3 and Mettl14 inactivation in late RPCs.

“Overexpression of Zfp292 phenocopied the effects of Mettl3 and Mettl14 inactivation in late RPCs (Xin et al., 2022; Li et al., 2023).”

• cites Xin et al., 2022 sourcePartly supported

  The Results section reports that Zfp292 overexpression delayed cell-cycle exit and increased Müller glial fate in late RPCs, effects that overlap with those reported for Mettl3 deficiency. But the Discussion explicitly says overexpression did not faithfully recapitulate the Mettl3-knockout cell-population phenotype, and the passages found do not establish the claimed comparison with Mettl14 inactivation. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites Li et al., 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

METTL3 is observed predominantly in nuclear speckles.

“METTL3 is observed predominantly in nuclear speckles, although m6A methyltransferase activity has been detected in both nuclear and cytoplasmic fractions, including in RPCs (Lin et al., 2016).”

• cites Lin et al., 2016 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Histone clusters served as a quality control for eCLIP assay and have been previously reported target RNAs for METTL3.

“We identified histone clusters, which served as a quality control for eCLIP assay, which have been previously reported target RNAs for METTL3 (Figure S4A) (Van Nostrand et al., 2016).”

• cites Van Nostrand et al., 2016 sourcePartly supported

  In Results, ‘Input normalization improves CLIP signal-to-noise’ (page not provided), the paper uses SLBP binding to histone RNAs to distinguish true eCLIP signal from false-positive signal, supporting their use as an assay check. The cited paper does not identify histone RNAs as previously reported METTL3 targets; that part of the claim is unsupported by our copy. (LLM, conf=0.87)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

m6A modifications reduce RNA stability.

“Classically, m6A modifications are thought to reduce RNA stability; thus loss of m6A leads to increased expression of corresponding RNA (Wang et al., 2014).”

• cites Wang et al., 2014 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

m6A methyltransferase activity has been detected in both nuclear and cytoplasmic fractions, including in RPCs.

“METTL3 is observed predominantly in nuclear speckles, although m6A methyltransferase activity has been detected in both nuclear and cytoplasmic fractions, including in RPCs (Lin et al., 2016).”

• cites Lin et al., 2016 sourcePartly supported

  The Introduction (page not provided) says that m6A methyltransferase activity has been detected in both nuclear and cytoplasmic extracts. It does not say that this activity was detected in RPCs; searching the indexed copy for “RPCs” found no supporting passage. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

Loss of m6A leads to increased expression of corresponding RNA.

“Classically, m6A modifications are thought to reduce RNA stability; thus loss of m6A leads to increased expression of corresponding RNA (Wang et al., 2014).”

• cites Wang et al., 2014 sourceSupported

Classified as probably not a substantial claim. May combine two claims.

The YTHDF family consists of YTHDF1, 2, and 3, and these three m6A readers have overlapping roles in mESCs.

“The YTHDF family consists of YTHDF1, 2, and 3, and these three m6A readers have overlapping roles in mESCs (Lasman et al., 2020).”

• cites Lasman et al., 2020 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

m6A depletion via 3′UTR mutations of insulin mRNA or indirectly by mutating Mettl3 decreased insulin protein production, leading to aberrant homeostasis and diabetic-like phenotype.

“Interestingly, m6A depletion via 3′UTR mutations of insulin mRNA or indirectly by mutating Mettl3 decreased insulin protein production, leading to aberrant homeostasis and diabetic-like phenotype (Wilinski and Dus 2023).”

• cites Wilinski and Dus 2023 sourceSupported

May combine two claims.

A functional nuclear localization signal (NLS) on METTL3 has been reported, and NLS mutant of METTL3 retains in cytoplasm.

“A functional nuclear localization signal (NLS) on METTL3 has been reported, and NLS mutant of METTL3 retains in cytoplasm (Schöller et al., 2018).”

• cites Schöller et al., 2018 sourceSupported

Classified as probably not a substantial claim.

The dTAG system (Nabet et al., 2018) was adapted to induce FKBP12F36V-METTL3 acute degradation.

“The dTAG system (Nabet et al., 2018) was adapted to induce FKBP12F36V-METTL3 acute degradation.”

• cites Nabet et al., 2018 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Several reports identified specific crosstalk between m6A RNA modifications and histone modifications, including H3K36me3, H3K9me2, H3K9me3, and K3K27Ac.

“Indeed, several reports identified specific crosstalk between m6A RNA modifications and histone modifications, including H3K36me3 (Huang et al., 2019), H3K9me2 (Li et al., 2020a), H3K9me3 (Liu et al., 2021a; 2021b), and K3K27Ac (Huang et al., 2025).”

• cites Huang et al., 2019 sourceSupported
• cites Li et al., 2020aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Liu et al., 2021aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

m6A modifications in the 3′UTR are recognized by YTHDF2 to regulate mRNA degradation.

“m6A modifications in the 3′UTR are recognized by YTHDF2 to regulate mRNA degradation (Wang et al., 2014), while synergistic interactions between YTHDF1 and YTHDF3 allow recognition of m6A in the 3′UTR to facilitate translation (Wang and Roundtree, 2015; Shi et al., 2017).”

• cites Wang et al., 2014 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Synergistic interactions between YTHDF1 and YTHDF3 allow recognition of m6A in the 3′UTR to facilitate translation.

“m6A modifications in the 3′UTR are recognized by YTHDF2 to regulate mRNA degradation (Wang et al., 2014), while synergistic interactions between YTHDF1 and YTHDF3 allow recognition of m6A in the 3′UTR to facilitate translation (Wang and Roundtree, 2015; Shi et al., 2017).”

• cites Wang and Roundtree, 2015 sourcePartly supported

  The Results sections “YTHDF1 Binds m6A inside Cells” and “YTHDF1 Modulates the Translation Dynamics of m6A-Modified mRNA” show that YTHDF1 recognizes m6A and promotes translation. The Discussion places YTHDF1-binding sites near the stop codon and in the 3′ UTR. The cited paper does not establish a synergistic interaction between YTHDF1 and YTHDF3; its discussion of cooperation concerns YTHDF1 and YTHDF2. Page numbers were not available in our copy. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites Shi et al., 2017 sourcePartly supported

  The ‘YTHDF3 facilitates translation’ section reports that YTHDF3 amplified YTHDF1’s translation-promoting effect in a 3′-UTR tethering assay. But that assay replaced the proteins’ m6A-binding domains with artificial tethering domains, so it does not directly show that their interaction recognizes m6A in the 3′UTR. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

For Dex inducible WT or APPA Mettl3 plasmid, GRBD insert from pPyCAG-cGR-IP plasmid (RIKEN, #RDB10440) was subcloned to constitutive PiggyBac plasmid to generate GRBD-FLAG-METTL3WT or GRBD-FLAG-METTL3APPA fusion protein (Murakami et al., 2016).

“For Dex inducible WT or APPA Mettl3 plasmid, GRBD insert from pPyCAG-cGR-IP plasmid (RIKEN, #RDB10440) was subcloned to constitutive PiggyBac plasmid to generate GRBD-FLAG-METTL3WT or GRBD-FLAG-METTL3APPA fusion protein (Murakami et al., 2016).”

• cites Murakami et al., 2016 sourceCould not verify

  The Methods section on the PiggyBac Tet-on system describes doxycycline-inducible constructs for genes including Nanog, not the claimed dexamethasone-inducible METTL3 constructs. Searches for GRBD, pPyCAG-cGR-IP, METTL3 and APPA found no passage describing the specified insert, subcloning step or fusion proteins in our copy. (LLM, conf=0.75)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

YTHDF1/2/3 are redundant in their roles to promote degradation of m6A-modified RNAs.

“YTHDF1/2/3 are redundant in their roles to promote degradation of m6A-modified RNAs (Zaccara and Jaffrey 2020, 2024; Niu et al., 2022).”

• cites Zaccara and Jaffrey 2020, 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Niu et al., 2022 sourcePartly supported

  The Discussion (page not available in our copy) says YTHDF1/2/3 regulate degradation and translation of target mRNAs, and reports that they function redundantly in retinal neurogenesis. It does not establish that the three readers are redundant specifically in promoting degradation of m6A-modified RNAs. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

A recent report demonstrated crosstalk between m6A RNA methylation and 5mC DNA methylation by METTL14-mediated recruitment of DNMT1 to chromatin, and the balance between these two marks controls key differentiation genes in mESCs.

“A recent report demonstrated crosstalk between m6A RNA methylation and 5mC DNA methylation by METTL14-mediated recruitment of DNMT1 to chromatin, and the balance between these two marks controls key differentiation genes in mESCs (Quarto et al., 2025).”

• cites Quarto et al., 2025 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Some m6A-modified chromatin-associated RNAs can bind directly to DNA to regulate chromatin state and downstream transcription.

“Some m6A-modified chromatin-associated RNAs can bind directly to DNA to regulate chromatin state and downstream transcription (Liu et al., 2020).”

• cites Liu et al., 2020 sourcePartly supported

  The paper's Abstract reports that m6A-modified chromosome-associated regulatory RNAs affect chromatin state and downstream transcription. It does not establish the claim that these RNAs bind directly to DNA; the passages on chromatin-associated RNAs and recruited proteins do not demonstrate direct RNA–DNA binding. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

METTL3-eCLIP was performed as described previously (Van Nostrand et al., 2016).

“METTL3-eCLIP was performed as described previously (Van Nostrand et al., 2016).”

• cites Van Nostrand et al., 2016 sourceSupported

May combine two claims.

METTL3 binds RPC chromatin and regulates bivalent histone modifications and chromatin accessibility, but these events do not alter transcription.

“We observe that METTL3 binds RPC chromatin and regulates bivalent histone modifications and chromatin accessibility, but these events do not alter transcription.”

May combine two claims.

Mettl3 loss impairs induction of EFTF genes Lhx2, Six3, and Rax but not loss of pluripotency or induction of Sox2/Otx2/Pax6 neuroectodermal genes.

“In both wild-type (WT) and Mettl3KO organoids, we observed comparable loss of pluripotency gene Nanog as well as similar induction of Sox2+/Otx2+/Pax6+ neuroectodermal genes. In contrast, the induction of EFTF genes Lhx2, Six3, and Rax was remarkably impaired”

May combine two claims.

Neither nuclear METTL3APPA nor cytosolic METTL3APPA rescues RPC differentiation defects.

“Neither nucleic METTL3APPA (+Dex) nor cytosolic METTL3APPA (-Dex; -Dox) could rescue the defects (Figures 1F and S1F).”

May combine two claims.

Down-regulated genes in Mettl3 KO vs WT RPCs are less enriched with H3K4me3 and harbor H3K27me3 marks.

“we found that the down-regulated genes in Mettl3 knockout (KO) vs. WT RPCs were indeed less enriched with H3K4me3 modifications, but these sites harbored H3K27me3 marks (Figure 2B).”

The quote only partly states this claim.

m6A modifications are enriched at the 3′UTR in WT and Mettl3KO organoids.

“We analyzed distribution of m6A modifications across RNA domains, including enrichment at the 3′UTR (Figure 3C)”

May combine two claims.

Bulk RNA-seq showed DEGs in Mettl3KO organoids were rescued by nucleus-translocated METTL3 (+Dex), but not cytosolic METTL3 (-Dex).

“Bulk RNA-seq showed differentially expressed genes in Mettl3KO organoids were rescued by nucleus-translocated METTL3 (+Dex), but not cytosolic METTL3 (-Dex).”

May combine two claims.

Cytosol-to-nuclear METTL3 translocation induced by Dex rescued Rx:GFP+ expression in Mettl3KO RPCs, but cytosolic METTL3 did not.

“Cytosol-to-nuclear METTL3 translocation was induced by adding Dex (10 μM), which rescued the Rx:GFP+ expression in Mettl3KO RPCs, but cytosolic METTL3 (without Dex, without Dox) did not.”

May combine two claims.

The most enriched GO pathway among METTL3-bound RNAs is RNA binding (p = 2.60e−44), followed by ribonucleoprotein complex and mRNA processing.

“Gene Ontology (GO) term pathway analysis of the 2584 METTL3-bound RNAs showed that the most enriched pathway is RNA binding (p = 2.60e−44) and followed by ribonucleoprotein complex and mRNA processing (Figure 4B).”

May combine two claims.

Acute degradation of endogenous METTL3 by dTAG-13 is near-complete as quickly as 2 h post-treatment or as low as 62.5 nM.

“near-complete degradation of FKBP12F36V-HA-METTL3 as quickly as 2 h post-dTAG-13, or as low as 62.5 nM of dTAG-13, was observed (Figure S4B).”

May combine two claims.

Mettl3-dependent H3K4me3 marks are depleted of H3K4me3 and enriched with H3K27me3 in Mettl3KO RPCs.

“We observed Mettl3-dependent H3K4me3 marks that were not only depleted of this modification in Mettl3KO RPCs but enriched with the repressive H3K27me3 modification at these sites (Figure 2A).”

May combine two claims.

Lhx2 carries m6A modifications both at 5′UTR and 3′UTR, Six3 at 3′UTR, and Rax at 5′UTR.

“Lhx2 carries m6A modifications both at 5′UTR and at 3′UTR, Six3 with m6A at 3′UTR and Rax at 5′UTR (Figure S3B).”

May combine two claims.

WT and Mettl3KO organoids harbor 28,689 and 11,881 high-confidence m6A sites, respectively.

“we found that WT and Mettl3KO organoids harbor 28,689 and 11,881 high-confidence m6A sites, respectively.”

May combine two claims.

Following acute METTL3 degradation, there were changes in transcription and m6A modifications.

“While we saw changes in transcription and m6A modifications, we did not observe major alterations to chromatin accessibility, indicating an uncoupling between the epigenome and epitranscriptome in this context.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

WDR5 controls EFTF activity and cell fate during retinogenesis.

“we found that WDR5, a histone methyltransferase co-factor, controls EFTF activity and cell fate during retinogenesis (Li et al. 2020b, 2021).”

• cites Li et al. 2020b, 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

In zebrafish, Mettl3 knockdown leads to microphthalmia due to delayed RPC differentiation and increased cell death.

“In zebrafish, Mettl3 knockdown leads to microphthalmia due to delayed retinal progenitor cell (RPC) differentiation and increased cell death (Huang et al., 2021).”

• cites Huang et al., 2021 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

METTL3 non-canonically regulates chromatin accessibility in mESCs via m6A-dependent interaction with histone-modifying enzymes and chromatin-regulatory RNAs, and in fibroblasts through m6A-independent association with enhancers.

“METTL3 non-canonically regulates chromatin accessibility in mESCs via an m6A-dependent interaction with histone-modifying enzymes and chromatin-regulatory RNAs, as well as the senescence-associated secretary phenotype in fibroblasts through m6A-independent association with enhancers (Liu et al. 2020, 2021a; Xu et al., 2021).”

• cites Liu et al. 2020, 2021aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Xu et al., 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

METTL3-dependent m6A modifications have been implicated in regulation of chromatin accessibility.

“Since METTL3-dependent m6A modifications have been implicated in regulation of chromatin accessibility (Deng et al., 2022)”

• cites Deng et al., 2022 sourceNot checked — no full text of the source

May combine two claims.

METTL3 regulates histone modifications, but its direct targets on chromatin do not contribute to gene regulation during mESC-to-RPC differentiation.

“METTL3 regulates histone modifications, but its direct targets on chromatin do not contribute to gene regulation during mESC-to-RPC differentiation”

Probably not a substantial claim: may not pass S3 (load-bearing).

Bivalent chromatin dynamics accompanies early mESC lineage commitment.

“Moreover, bivalent chromatin dynamics accompanies early mESC lineage commitment (Ruthenburg et al., 2007; Schwartz and Pirrotta 2007; Liu et al., 2016).”

• cites Ruthenburg et al., 2007 sourceNot checked — no full text of the source
• cites Schwartz and Pirrotta 2007 sourceNot checked — no full text of the source
• cites Liu et al., 2016 sourceNot checked — no full text of the source

May combine two claims.

METTL3 associates with specific H3K4me3 modifications in Mettl3-dependent DEGs, likely related to activation of specific genes during mESC-to-RPC differentiation.

“Our data indicate that METTL3 associates with specific H3K4me3 modifications in Mettl3-dependent DEGs, likely related to activation of specific genes during the mESC-to-RPC differentiation program.”

May combine two claims.

METTL3 regulates the mESC-to-RPC and RPC-to-retina transcriptional programs.

“Here, we show that METTL3, the main m6A writer, regulates the mESC-to-RPC and RPC-to-retina transcriptional programs through mechanisms independent of chromatin accessibility and its ability to directly engage chromatin.”

• AssumptionRests on an assumption: METTL3 is the main m6A writer.

May combine two claims.

METTL3 chromatin targets are not functional: not associated with changes in bivalent histone modifications or transcript expression in RPCs.

“we conclude that METTL3 chromatin binding in RPCs is uncommon and that METTL3 chromatin targets are not functional: they are not associated with changes in bivalent histone modifications or transcript expression in RPCs.”

May combine two claims.

Targeted acute METTL3 loss in RPCs leads to impaired retinal development in vitro and in vivo.

“Targeted acute METTL3 loss in RPCs leads to impaired retinal development in vitro and in vivo”

Probably not a substantial claim: may not pass S3 (load-bearing).

More than 95% transduction efficacy was seen on day 6 Rx:GFP+ RPCs following day 0 infection.

“More than 95% transduction efficacy was seen on day 6 Rx:GFP+ RPCs following day 0 (mESC) infection (Figure 3F).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Two independent Mettl3 guide RNAs were used to generate independent Mettl3 KO ES clones.

“For CRISPR-Cas9-mediated Mettl3 knockout, two independent Mettl3 guide RNAs (GTGCTTATTGATAATTCGTCTGG and TGTGAAGCTGGTTAAAGCTCAGG) were used to generate independent Mettl3 KO ES clones.”

Classified as probably not a substantial claim.

Both the dose-dependent assay and time-course assay of dTAG-13 treatment using the knockin Rx:GFP mESC cell line were performed to determine the ideal dose and time point of dTAG-13 treatment to achieve acute METTL3 degradation.

“Both the dose-dependent assay and time-course assay of dTAG-13 treatment using the knockin Rx:GFP mESC cell line were performed to determine the ideal dose and time point of dTAG-13 treatment to achieve acute METTL3 degradation.”

Classified as probably not a substantial claim.

The knockin stable colonies were screened by genotyping and western blot with both HA antibody and METTL3 antibody.

“The knockin stable colonies were screened by genotyping and western blot with both HA antibody and METTL3 antibody.”

Classified as probably not a substantial claim.

The detection of m6A at targeted sites was based on the SELECT method previously developed by Xiao et al. (2018).

“The detection of m6A at targeted sites was based on SELECT method previously developed by Xiao et al., (2018).”

• cites Xiao et al., (2018) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

CUT&RUN was performed as previously described (Li et al., 2020b; Li et al. 2020b; Li et al. 2021).

“2×106 cells were subjected to CUT&RUN per antibody as previously described (Li et al., 2020b; Li et al. 2020b; Li et al. 2021).”

• cites Li et al., 2020bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Li et al. 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

ChIP-seq was performed using a slightly modified version of ChIP Assay Kit (Millipore)-based protocol as described previously (Li et al. 2020b, 2021).

“ChIP-seq was performed using a slightly modified version of ChIP Assay Kit (Millipore)-based protocol as described previously (Li et al. 2020b, 2021).”

• cites Li et al. 2020b, 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Two biological replicates were conducted for each experiment.

“Two biological replicates were conducted for each experiment.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Rx:GFP ESC maintenance, induced retinal neuroectoderm differentiation, and plasmid electroporation were performed as described previously (Eiraku et al., 2011; Li et al. 2020b, 2021).

“Rx:GFP ESC maintenance, induced retinal neuroectoderm differentiation, and plasmid electroporation were performed as described previously (Eiraku et al., 2011; Li et al. 2020b, 2021).”

• cites Eiraku et al., 2011 sourceNot checked — no full text of the source
• cites Li et al. 2020b, 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

WT or mutant Mettl3 inserts (APPA, W475A and N477A) were subcloned to Dox-inducible or constitutive PiggyBac plasmids as described previously (Xu et al., 2021).

“WT or mutant Mettl3 inserts (APPA, W475A and N477A) were subcloned to Dox-inducible or constitutive PiggyBac plasmids as described previously (Xu et al., 2021).”

• cites Xu et al., 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

GLORI was performed as reported previously (Liu et al., 2023), slightly modified.

“GLORI was performed as reported previously (Liu et al., 2023), slightly modified, as follows:”

• cites Liu et al., 2023 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

We cannot rule out an indirect effect of METTL3 on chromatin modification.

“While we did not find many METTL3-binding sites on chromatin, and that METTL3 chromatin targets did not show differential transcription, we cannot rule out an indirect effect of METTL3 on chromatin modification, given that we observed METTL3-dependent changes in histone modifications.”