Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum

Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum Yunjia Zhang , Youning Lin , Xunan Shen , Xue Xiao , Cai Song , Xiaobo Shi , Tao Zhou , Yanxin Li , Zihan Wu , Zhenkun Zhuang , Chunqiong Li , Meng Li , Feng Wen , Jianlin Liu , Qiangqiang Zhang , Zhao-Lu Li , Songbo Zhang , Lei Cao , Susu Qu , Yaqi Li , Jianhua Yao , Fubaoqian Huang , Xin Liu , Ziqing Deng , Longqi Liu , Xun Xu , Jianwei Jiao , Li Zhang , Shiping Liu , Yaoyao Zhang doi: https://doi.org/10.1101/2025.08.14.670213 Yunjia Zhang 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Youning Lin 2 BGI Research , Hangzhou 310030, China 3 BGI Research , Shenzhen 518083, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Xunan Shen 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Xue Xiao 4 Department of Obstetrics and Gynecology, Key Laboratory of Birth Defects and Related of Women and Children of Ministry of Education, West China Second University Hospital, Sichuan University, Frontier Medical Center, Tianfu Jincheng Laboratory , Chengdu, Sichuan 610041, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Cai Song 2 BGI Research , Hangzhou 310030, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Xiaobo Shi 3 BGI Research , Shenzhen 518083, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Tao Zhou 2 BGI Research , Hangzhou 310030, China 5 College of Life Sciences, University of Chinese Academy of Sciences , Beijing 100049, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Yanxin Li 6 State Key Laboratory of Organ Regeneration and Reconstruction, Institute of Zoology, Chinese Academy of Sciences , Beijing, 100101, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zihan Wu 7 AI for Life Sciences Lab , Tencent, Shenzhen 518057, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zhenkun Zhuang 2 BGI Research , Hangzhou 310030, China 3 BGI Research , Shenzhen 518083, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Chunqiong Li 8 Chinese Institute for Brain Research , Beijing 102206, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Meng Li 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Feng Wen 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jianlin Liu 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Qiangqiang Zhang 9 Advanced Model Animal Research Center, Department of Biotechnology and Biomedicine, Yangtze Delta Region Institute of Tsinghua University , Zhejiang 314006, China 10 Zhejiang Key Laboratory of Multiomics and Molecular Enzymology, Yangtze Delta Region Institute, Tsinghua University , Zhejiang 314006, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zhao-Lu Li 11 IDG/McGovern Institute for Brain Research, Tsinghua-Peking Center for Life Sciences, School of Life Sciences, Tsinghua University , Beijing 100084, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Songbo Zhang 11 IDG/McGovern Institute for Brain Research, Tsinghua-Peking Center for Life Sciences, School of Life Sciences, Tsinghua University , Beijing 100084, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Lei Cao 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Susu Qu 8 Chinese Institute for Brain Research , Beijing 102206, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Yaqi Li 12 Peking Union Medical College , Beijing 100730, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jianhua Yao 7 AI for Life Sciences Lab , Tencent, Shenzhen 518057, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Fubaoqian Huang 2 BGI Research , Hangzhou 310030, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Xin Liu 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Ziqing Deng 1 BGI Research , Beijing 102601, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Longqi Liu 2 BGI Research , Hangzhou 310030, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Xun Xu 2 BGI Research , Hangzhou 310030, China 3 BGI Research , Shenzhen 518083, China 14 State Key Laboratory of Genome and Multi-omics Technologies, BGI Research , Hangzhou 310030, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jianwei Jiao 6 State Key Laboratory of Organ Regeneration and Reconstruction, Institute of Zoology, Chinese Academy of Sciences , Beijing, 100101, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jwjiao{at}ioz.ac.cn zhangli{at}cibr.ac.cn liushiping{at}genomics.cn yaoyaozhang{at}scu.edu.cn Li Zhang 8 Chinese Institute for Brain Research , Beijing 102206, China 13 Institute of Blood Transfusion, Chinese Academy of Medical Sciences , Chengdu 610052, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jwjiao{at}ioz.ac.cn zhangli{at}cibr.ac.cn liushiping{at}genomics.cn yaoyaozhang{at}scu.edu.cn Shiping Liu 2 BGI Research , Hangzhou 310030, China 14 State Key Laboratory of Genome and Multi-omics Technologies, BGI Research , Hangzhou 310030, China 15 Key Laboratory of Spatial Omics of Zhejiang Province, BGI Research , Hangzhou 310030, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jwjiao{at}ioz.ac.cn zhangli{at}cibr.ac.cn liushiping{at}genomics.cn yaoyaozhang{at}scu.edu.cn Yaoyao Zhang 4 Department of Obstetrics and Gynecology, Key Laboratory of Birth Defects and Related of Women and Children of Ministry of Education, West China Second University Hospital, Sichuan University, Frontier Medical Center, Tianfu Jincheng Laboratory , Chengdu, Sichuan 610041, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jwjiao{at}ioz.ac.cn zhangli{at}cibr.ac.cn liushiping{at}genomics.cn yaoyaozhang{at}scu.edu.cn Abstract Full Text Info/History Metrics Supplementary material Preview PDF Summary The human fetal brain undergoes morphological changes that contribute to the development of regional functionalities. However, the features of structural development, the underlying molecular and cellular signatures in the fetal brain remain unclear. With spatial transcriptomics and snRNA-seq, we identified 25 forebrain regions and characterized the dynamic changes in the cortex and striatum during the late first and early second trimesters. In particular, we discovered that temporal lobe enriched NPY-expressing L2/3 EX neuron potentially interacted with L4 EX neurons during cortical expansion and arealization. Additionally, the gyrus and sulcus were developmental asynchronous, in which HOPX and SPARC genes were potentially involved. Further investigation on the striatum showed specific genes and cell types that enriched in patch and matrix compartments, and SST -positive interneurons potentially involved in the development of these structures. Together, our results give insights into the understanding of early fetal brain development. Introduction The fetal brain undergoes dynamic morphological and cellular alterations during development, resulting in a mature configuration of the adult brain, especially for the most evolved neocortex with convoluted, layered gray matter as well as the striatum related to cognitive control function 1 , 2 . During corticogenesis, neurons assemble in an inside-out layering manner. The laminar structure expands and undergoes transient changes, accompanied by fine-tuned layering and cortical arealization 3 . However, the precise spatiotemporal distribution of cell types across cortical regions during development at single_cell resolution hasn’t been resolved yet. Furthermore, whether distinct neuronal subtypes emerge in a region-specific manner during corticogenesis is unknown. Another prominent and evolutionary feature of the human brain is the folding of the neocortex 4 . The complex folding structure, which can be measured as a high gyrification index 5 , provides the possibility for a correlation with human-specific lobes. Diverse cellular and genetic mechanisms contribute to the folding structures. However, how convoluted structures systematically form in the developing human brain is unclear. These mentioned several unknowns in the cortex were the aims of the current study. The cortex is widely connected with the basal ganglia via segregated populations of striatal projection neurons 6 . Over 90% of striatal projection neurons are striatal medium spiny neurons (MSNs) 7 . Previous studies have shown the development of human MSNs in the proliferating zone of the lateral ganglionic eminence (LGE), where decoded TFs and lincRNAs regulate the differentiation of preMSNs into D1 and D2 MSNs 8 . However, the development and distribution of D1/D2 hybrid MSNs remain unknown. In the mantle zone, MSNs and interneurons develop into a nucleus consisting of patch and matrix compartments in the basal ganglionic system. Studies on radiolabeled postmitotic neurons in rat embryos showed that patch neurons were born earlier than matrix neurons, and therefore established connections with the substantia nigra earlier 9 , 10 . While to date, the cellular properties involved in the development of patches and matrices, as well as the molecular dynamics driving structural identities are unclear. Thus, these two unclears in the striatum are the other aims of the current study. To address these questions on nonpathological human fetal tissue, a combination of spatiotemporal seq and snRNA seq was used in the present study, which provides an opportunity to understand the structural development and changes of molecular and cellular signatures in the cortex and striatum. This resource is interactively accessible at: http://db.cngb.org/cnsa/project/CNP0005272_8c0dc42f/reviewlink/ Results Spatially resolved cellular map of the human fetal brain To comprehensively characterize the dynamics of spatiotemporal gene expression dynamics in developing human fetal brains, we employed an integrated multi-omics approach combining high-resolution spatial transcriptomics (stereo-seq 11 ) and single-nucleus RNA sequencing (snRNA-seq; DNBelab C4 11 ). The spatial profile was conducted on coronal sections across cortical and striatal regions at gestational week (GW) 11,15, 21, and included data from Li et.al 12 , resulting in spatial data spanning GW8 to GW21, and encompassing first trimester (GW8, 10-12), and second trimester (GW15-16, 21) developmental stages. Single cell resolution was achieved by a cell segmentation algorithm 13 . Parallel snRNA-seq analysis was performed on histologically matched anatomical sections (coronal) to resolve cell-type-specific transcriptional signatures at corresponding developmental timepoints ( Fig 1A ). A total of 424,527 cells were generated by snRNA-seq. After integrated the published data 12 , 14 , 15 , we finally approached 798,138 cells for the human fetal brain. All the cells were clustered into 12 major cell types and 136 subtypes ( Figs 1B , S1A-D ). Download figure Open in new tab Fig.S1 | Quality control, major cell types and subtypes of snRNA-seq data, related to Fig.1 . A. Quality control of snRNA-seq clusters by counts and RNA features. B. Stack-bar plot showing the percentage of major cell types in each library and sample. C . Bubble plot showing the expression levels of the specific markers in each of the major cell types. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. D. MILO analysis showing the distribution of 134 (out of 136 cell subtypes, NPC_EX4 and IN13 were removed due to low MILO value) cell subtypes along developmental stages. The spatial brain regions in each coronal section were delineated using FuseMap 16 at single cell resolution, as well as SpaGCN clustering 17 at bin100 resolution, leading to an identification of 25 subregions within the developing brain, which were confirmed through the canonical marker genes ( Figs 1C, 1D , S2A-F , S3A, S3B ). The cortex was consisted of the ventricle zone (VZ) ( VIM, SOX2 ) 18 , subventricular zone (SVZ) ( PPP1R17, EOMES ) 19 , intermediate zone (IZ, identified based on anatomy structure), subplate (SP) ( ST18, NR4A2 ) 20 and cortical plate (CP) ( NEUROD2, TBR1 ) 19 ( Figs 1C, 1D , S2B, S2C, S2F ). The subpallial regions developed the hippocampus (HIP) ( NEUROD1, NR3C2) 21 , 22 , GE-proliferating zone ( GAD2 ) 15 , LGE-proliferating zone ( PAX6, SIX3 ) 15 , LGE-mantle zone (also promising striatum, STR) ( TAC1, EBF1 ) 15 , MGE-proliferating zone ( NKX2-1 ) 15 , CGE-proliferating zone ( NR2F2 ) 15 , globus pallidus (GP) ( LHX8, LHX6 ) 23 , amygdala (AMY) ( NRP2, NR2F2 ) 24 , septum (SEP) ( ZIC4, ZIC1, ZIC2 ) 25 , 26 , claustrum (Cla) ( NTNG2, NR4A2 ) 27 , 28 , thalamus (THT) ( TCF7L2, LHX9 ) 29 , hypothalamus (HYT) ( LHX5-AS1 ) 30 , bed nucleus of striatal terminal (BNST) (C CK, NTS, SST, ADRA1B ) 31 and etc. ( Figs 1C, 1D , S2B, S2C, S2F , S3B ). Brain regions in sagittal sections from Li et.al were spatially clustered with SpaGCN at bin100 ( Figs S3C-E ), followed by single-cell resolution annotation ( Figs S4A, S4B ). The spatially resolved regions within the sagittal plane exhibited significant spatial congruence with established anatomical boundaries ( Fig S4C ). Both coronal and sagittal brain regions exhibited strong correlation across different developmental stages ( Fig S4D ). By single-cell data integration with spatial transcriptomic mapping, we identified the cerebral cortex as the regions with the most cell number across the sampled regions ( Fig 1E ). Download figure Open in new tab Fig.S2 | Quality control, spatial region registration of stereo-seq data, related to Fig.1 . A. Box plot showing the number of detected UMIs per brain region on spatial transcriptomic data, clustered by FuseMap at single cell resolution. B. Bubble plot showing the expression levels of the specific markers in each of the sub-brain regions, clustered by FuseMap at single cell resolution. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. C. Spatial clustering of stereo-seq brain regions by SpaGCN at bin100 resolution. Brain regions are colored by the annotations. VZ: ventricle zone; SVZ: sub-ventricle zone; ISVZ: inner sub-ventricle zone; OSVZ: outer sub-ventricle zone; IZ: intermediate zone; SP: subplate; CP: cortical plate; LGE: lateral ganglionic eminence; MGE: medial ganglionic eminence; CGE: caudal ganglionic eminence; STR: striatum (caudate and putamen); IC: internal capsule; AMY: amygdala; GP: globus pallidus; HTH: hypothalamus; THM: thalamus; SEP: septum; HIP: hippocampus; Pir: piriform cortex; EC: entorhinal cortex; Cla: claustrum; BNST: bed nucleus of striata terminal; NA: not available. D: dorsal; L: lateral. Scales: 5mm. D. Identification of ISVZ and OSVZ regions as well as key DEGs expression. Spatial subclustering of stereo-seq SVZ regions by SpaGCN at bin100 resolution. SVZ subregions are colored by the annotations. ISVZ: inner subventricular zone; OSVZ: outer subventricular zone. D: dorsal; L: lateral. Scales: 5 mm (Left). Bubble plot showing the expression levels of the top 5 DEGs enrichment in ISVZ and OSVZ. Bubble size shows the percentage of square bins expressed; the scale bar shows the average expression level (Right). E. Box plot showing the number of detected UMIs per brain region on spatial transcriptomic data, clustered by SpaGCN at bin100 resolution. F. Bubble plot showing the expression levels of the specific markers in each of the sub-brain regions, clustered by SpaGCN at bin100 resolution. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. Download figure Open in new tab Fig.S3 | Quality control, spatial region registration of stereo-seq data, related to Fig.1 . A. Heatmap illustrating the comparison of spatially defined brain subregions clustered by FuseMap and SpaGCN. B. Spatial visualization of specific brain region markers. White solid lines represent the borders of subregions. HIP: NEUROD1 , NR3C2 ; LGE-proliferating zone: PAX6 , SIX3 ; MGE-proliferating zone: NKX2-1 , CGE-proliferating zone: NR2F2 ; BNST: CCK , NTS , SST , ADRA1B . White dash lines represent the higher-magnification view of selected area. D: dorsal; L: lateral. Scales: 1mm. C. Spatial clustering of stereo-seq brain regions from Li’s data by SpaGCN at bin100 resolution. Brain regions are colored by the annotations. VZ: ventricle zone; SVZ: sub-ventricle zone; IZ: intermediate zone; SP: subplate; CP: cortical plate; GE: ganglionic eminence; BG: basal ganglia; IC: internal capsule; THM: thalamus; HTH: hypothalamus; CB: cerebellum; NA: not available. D: dorsal; A: anterior. Scales: 5mm. D. Box plot showing the number of detected UMIs per brain region on spatial transcriptomic data from Li et. al, clustered by SpaGCN at bin100 resolution. E. Bubble plot showing the expression levels of the specific markers in each of the brain regions from Li et. al, clustered by SpaGCN at bin100 resolution. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. Download figure Open in new tab Fig.S4 | Spatial region registration and comparison of stereo-seq data, related to Fig.1 . A. Spatial annotation of stereo-seq brain regions from Li’s data by SpaGCN at single cell resolution. Brain regions are colored by the annotations. VZ: ventricle zone; SVZ: sub-ventricle zone; IZ: intermediate zone; SP: subplate; CP: cortical plate; GE: ganglionic eminence; BG: basal ganglia; IC: internal capsule; THM: thalamus; HTH: hypothalamus; CB: cerebellum; NA: not available. D: dorsal; A: anterior. Scales: 5mm. B. Box plot showing the number of detected UMIs per brain region on spatial transcriptomic data from Li et. al, annotated by SpaGCN at single cell resolution. C. Heatmap illustrating the comparison of spatially defined brain subregions clustered by SpaGCN with anatomically defined brain regions from Li et. al. D. Sanky plot showing the correlation of brain regions across developmental stages. GW: samples from this study; gw: published data by Li et al. Download figure Open in new tab Fig.1 | Single-nuclei transcriptomics and stereo-seq characterize human fetal brain cellular map. A. Overall experimental design. Scales: 10μm (top); 1mm (bottom). GW: samples from this study; gw: published data by Li et al. B. UMAP showing snRNA-seq clustering and cell-type taxonomy. Cell types colored by the annotations (Left). EX: excitatory neuron; NPC_EX: neural progenitor cell_excitatory neuron; NPC_IN: neural progenitor cell_interneuron; IN: interneuron; EX_IN_MIX: excitatory neuron and interneuron mixed cells; APC/AST: astrocyte precursor cells/astrocyte; OPC/OL: oligodendrocyte precursor cells/oligodendrocyte; MIC: microglia; ENDO: endothelial cells; VLMC: vascular leptomeningeal cells. Cells colored by time points and data source (Right). C. Spatial clustering showing stereo-seq brain regions by FuseMap at single cell resolution. Brain regions are colored by the annotations. VZ: ventricle zone; SVZ: sub-ventricle zone; ISVZ: inner sub-ventricle zone; OSVZ: outer sub-ventricle zone; IZ: intermediate zone; SP: subplate; CP: cortical plate; LGE: lateral ganglionic eminence; MGE: medial ganglionic eminence; CGE: caudal ganglionic eminence; STR: striatum (caudate and putamen); IC: internal capsule; AMY: amygdala; GP: globus pallidus; HTH: hypothalamus; THM: thalamus; SEP: septum; HIP: hippocampus; Pir: piriform cortex; EC: entorhinal cortex; Cla: claustrum; BNST: bed nucleus of striata terminal; NA: not available. D: dorsal; L: lateral. Scales: 5mm. D. Spatial visualization showing specific brain region markers. White solid lines represent the borders of subregions. VZ: VIM , SVZ: PPP1R17 , SP: ST18 , CP: NEUROD2 , GE-proliferating zone: GAD2 , GP: LHX8 , AMY: NRP2 , SEP: ZIC4 , Cla: NTNG2 , THM: TCF7L2 , HTH: LHX5-AS1 . D: dorsal; L: lateral. Scales: 1mm. E. Stack-bar plot showing the distribution of clusters in brain regions. See also Figs. S1 , S2 , S3 , S4 . The dynamic distribution of excitatory cells in cortical expansion and arealization Spatial clustering analyses demonstrated that the cortical plate developed above the subplate from GW11 to gw16 ( Fig 1C , S2C , S3C , S4A ), aligning with the prevailing theory of prelate division into the subplate and cortical plate around PCW12 (equivalent to GW14) 32 . To quantitatively analyze spatiotemporal cellular dynamics during neocortical expansion, stratified proportions of cells within embryonic cortical laminae were quantified across neurodevelopmental stages ( Fig 2A ). The VZ and SVZ were mainly occupied by the NPC_EX, NPC_IN and APC/AST ( Fig 2A , S5A ). In the NPC_EX cluster, the NPC_EX16 primarily localized in the VZ, exhibiting RG characteristics with SOX2 expression, and NPC_EX2/7/1/6 predominantly resided in the SVZ and IZ, displaying IPC features with PPP1R17 expression. In contrast, no distinct RG- or IPC-only clusters were identified within the NPC_IN population, the NPC_IN17/14/15/16/7 expressed both SOX2 and ASCL1 , initially localized to the VZ during first trimester (GW8–12) and expanded into SVZ during the second trimester (GW15–21); while NPC_IN6/12/11 distributed through SVZ, IZ and the cortical plate across the neurodevelopmental stages ( Figs 2A , S5B, S5C ). Download figure Open in new tab Fig.S5 | The dynamic distribution of cells in cortical expansion and arealization, related to Fig.2 . A. Stack-bar plot showing the percentage of cells in each spatial cortical layer during development. B. Spatial plot showing 13 NPC cell types and glial cells at single cell resolution across different developmental stages. White solid lines represent the borders of sublayers. Yellow dotted boxes represent the zoom in region. D: dorsal; L: lateral. Scales:1mm. C. Bubble plot showing the expression levels of the specific markers in NPC clusters from snRNA data. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. D. Bubble plot showing the expression levels of the deep layer and upper layer markers in EX clusters from snRNA data. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. E. Spatial plot showing 14 excitatory cell types and 4 EX_IN_MIX cells at single cell resolution. White solid lines represent the borders of sublayers. D: dorsal; A: anterior. Scales:5mm. F. Spatial plot showing the distribution of 4 excitatory cell types in different cortical lobes at gw16. D: dorsal; A: anterior. Scales:5mm. G. Violin plot showing the expression level of the ligand NPY and the receptor NPY1R in snRNA data. H. Spatial visualization of the expression of the ligand NPY and the receptor NPY1R at GW21. D: dorsal; L: lateral. A dynamic spatiotemporal redistribution of cellular populations was observed within the SP and CP laminae across progressive stages of corticogenesis. The distribution of excitatory neurons segregated into three distinct populations: EX26/19/45/10 localized in the SP by GW12, co-expressing TLE4 and BCL11B , consistent with a deep layer neuron identity; EX5/29/30/2/11, early-born cortical neurons occupying SP/CP during the first trimester (GW8–12) and restricted to CP by the second trimester (GW15–21), highly expressing layer4 marker genes NECAB1, RORB ; EX42/33/22/32/20, late-born cortical neurons residing in VZ/SVZ/IZ during the first trimester, subsequently migrating to CP by GW15 with elevated expression of the upper-layer marker CUX2 and CUX1 ( Figs 2B, 2C , S5D, S5E ). In contrast to the increasement of excitatory cells in the SP and CP, the proportion of EX_IN_MIX cells dropped during cortical expansion ( Fig S5A ). Though they exhibited layer preference, only EX_IN_MIX4 located at CP by GW21, most of other cells were still in the IZ, revealing distinct developmental pattern in EX_IN_MIX cells ( Figs 2B, 2C ). At GW21, the cortical region developed into functional lobes. By calculating gene module via HotSpot 33 , we identified module 17 highly expressed in temporal lobe, and module 18 enriched in the frontal lobe ( Fig 2D ; Table S1 ). These lobe-enriched gene modules exhibited significant correlations with distinct EX subtypes ( Fig 2E ). Specifically, module 17 showed strong association with EX45, whereas module 18 was predominantly linked to EX11 ( Fig 2E ). To investigate the relations between cell type and lobes, we further examined the distribution of cells among different lobes. The excitatory neurons showed lobe specificity, with EX11 is enriched in dorsal lobe, EX33/22/10/45 preferentially distributed in the ventral cortex ( Figs 2F-G , S5F ). To study the function of these region enriched cells, the ligand-receptor interactions between cells in SP and CP were investigated. The NPY signaling pathway involving in food intake, circadian rhythms and memory 34 , 35 , 36 , abundant in GABAergic neurons, was specifically secreted by EX33 ( Figs 2H , S5G ). The NPY peptide and its receptor NPY1R both expressed in temporal cortex at GW21 ( Figs 2I , S5H ). The spatial adjacent localization of EX33 secreting NPY and EX22/EX2 expressing NPY1R , suggested the communication in excitatory neurons though NPY signaling pathways ( Fig 2J ). Moreover, EX33 and EX22 exhibited L2/3 neurons identify, while EX2 with L4 neuron feature ( Fig 2C ). Our results indicated that L2/3 EX neurons could modulate not only other L2/3 EX subclusters, but also the L4 EX subcluster (EX2) through NPY pathway in the developing temporal cortex. Download figure Open in new tab Fig.2 | The dynamic distribution of excitatory cells in cortical expansion and arealization. A. Pie chart showing the spatial distribution of cells in cortical subregions during development. GW: samples from this study; gw: published data by Li et al. B. Spatial mapping showing the distribution of 14 EX and 4 EX_IN_MIX cell types at single cell resolution at GW11, GW15 and GW21. White solid lines represent the borders of sublayers. D: dorsal; L: lateral. Scales:1mm. C. Heatmap showing the correlation between adult EX and spatially mapped fetal EX. D. Spatial visualization of lobe enriched gene modules using Hotspot. Scales: 1mm. E. Heatmap showing the correlation of gene modules and EX cell types. F. Pie chart showing the spatial distribution of cells in SP and CP of different lobes at GW21. G. Spatial plot showing EX cell types at single cell resolution in different lobes at GW21. White dotted box 1: frontal lobe; White dotted box 2: temporal lobe. D: dorsal; L: lateral. Scales:1mm. H. Chord plot showing NPY signaling pathway in snRNA using CellChat. I. Expression of NPY in ISH data at 15pcw and 21pcw from Allen Brain Atlas. Scales: 1mm. D: dorsal, V: ventral, L: lateral. J. Spatial visualization of the expression of ligand NPY in EX33 and the receptor NPY1R in EX22 and EX2 at GW21 at single cell resolution. Black box represents higher-magnification view. Scales: 1mm. D: dorsal, L: lateral. See also Fig.S5 and TableS1. The transcriptomic divergence of the gyrus and neighboring sulcus in the second trimester To investigate how structural patterning emerges during cortical development, we traced the folding of the neocortex across GW11, GW15, and GW21. The morphological characteristics of sulcal initiation was detected at GW15. At GW11, cortical cell differentiation was uniform along the tangential axis, whereas at GW15 and GW21, maturation trajectories extended from the sulcal to adjacent gyral regions, as revealed by velocity streamlines ( Fig 3A ). For further transcriptomic analysis, the gyri and sulci stereo-seq data were extracted at GW15 and GW21( Fig S6A, STAR Methods). A strong gene expression correlation (see STAR Methods ) was observed between the Sulcus_SP and Gyrus_SVZ ( r = 0.73), and between the Sulcus_IZ and Gyrus_SVZ ( r = 0.70) ( Fig 3B ). Regarding the germinal subregions, the Sulcus_SVZ also showed a strong correlation with the SVZ ( r = 0.77) and IZ regions of the Gyrus ( r = 0.66); the Sulcus_VZ showed highest correlation with Gyrus_SVZ( r = 0.64) rather than the Gyrus_VZ ( r = 0.53) ( Fig 3B ). These findings demonstrated distinct maturation gradients across cortical compartments: sulcal SP and IZ exhibited delayed maturation relative to their gyral counterparts, whereas sulcal VZ and SVZ mirrored the transcriptional profiles of gyral SVZ. Download figure Open in new tab Fig. 3 | Spatial heterogeneity of cell types in human cortical folding development. A. Pseudotime trajectory analysis at GW11, GW15 and GW21 via SpaTrack. Cortical regions are clustered by FuseMap. Black arrows denote the maturation trend, blue arrows indicate gyri, and red arrows signify sulci. Scales:1mm. B. Heatmap showing the gene expression module correspondence between sulcus (x axis) subregions and neighboring gyrus (y axis) subregions. Subregion clusters from FigS6A. C. Spatial distribution of 9 major cell types in an example cortex section for gyrus and sulcus at GW15. Scales: 500μm. C. Differential cell type proportions between cortical gyri and sulci. Cell types exceeding 1% abundance were shown. Individual points represent cell type-specific proportions. Bubble size corresponds to -log₁₀(adjusted P-value) (Chi-squared test with Benjamini-Hochberg correction). Diagonal line represents equal proportions (y = x). D. Stack-bar plots showing the percentage of cell subtypes in each cortical layer between gyrus and sulcus. E. MILO analysis showing the proportions of cell types between the gyrus and sulcus across GW11 to GW21. Cell types with statistically significant differential variation patterns (p-value < 0.05, Chi-squared test with Benjamini-Hochberg correction) are highlighted in blue. F. Differential expression of FTL , HOPX , PTN , and SPARC in selected cell types between gyrus (pink) and sulcus (cyan). Cell types analyzed: NPC_IN14, NPC_EX16, EX19, EX2, APC/AST8. Significance assessed by two-tailed t-test (*p<0.05, **p<0.01, ***p<0.001). G. Spatial expression of HOPX between gyrus and sulcus at GW15. Scales:1mm. H. Spatial expression of Hopx in mouse E16.5 cortex sagittal section ( MOSTA data) (Left). ISH staining of Hopx in mouse E15.5 cortex sagittal section ( Allen Atlas ) (Right). Scales:1mm. I. Spatial expression of SPARC between gyrus and sulcus at GW15. Scales:1mm. J. Spatial expression of Sparc in mouse E16.5 cortex sagittal section (MOSTA data) (Left). ISH staining of Sparc in mouse E15.5 cortex sagittal section (Allen Atlas) (Right). Scales:1mm. See also Fig.S6 and Table S2, S3. Download figure Open in new tab Fig.S6 | The occupied region, transcriptomic and the cellular heterogeneity in developing gyrus and sulcus, related to Fig.3 . A. Spatial clustering of lassoed gyrus and adjacent sulcus region stereo-seq coronal sections at GW15 and GW21 by FuseMap. Subregions are colored by the annotations: VZ (blue), SVZ (pink), IZ (green), SP (yellow), CP (light blue), Pial(purple). Scales: 500μm. B. Stack-bar plots showing the percentage of cell subtypes between gyrus and sulcus. (Chi-squared test with Benjamini-Hochberg correction, *p<0.05, **p<0.01, ***p<0.001). C. Volcano plots illustrate gyrus-sulcus comparisons in five neurodevelopmental subregions: VZ, SVZ, IZ, SP, and CP. Significantly DEGs are defined by Benjamini-Hochberg adjusted p < 0.05 (two-tailed Wald test). Color encoding: red (p < 0.05), navy blue (non-significant, p ≥ 0.05). D. GO terms for genes upregulated in gyrus (p-value: Fisher’s exact test, one-sided). E. Bubble plot illustrates the expression levels of seven DEGs ( FTL, HOPX, PTN, SPARC, ID2, ZEB1, VIM ) demonstrating significant gyral enrichment across snRNA-seq defined cortical cell clusters. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. F. Spatial expression of Id2 , Zeb1 and Vim in mouse E14.5 cortex sagittal section (MOSTA data). Scales: 1mm. G. Spatial expression of Id2 , Zeb1 and Vim in mouse E16.5 cortex sagittal section (MOSTA data). Scales: 1mm. To characterize spatial differences in cell type proportions between the gyrus and sulcus brain regions, we integrated snRNA-seq data with spatial transcriptomic data ( Fig 3C ) and quantitatively analyzed their regional distributions ( Figs 3D , S6B ). Overall, 27 specific cell types proportions in the gyrus different from those in the sulcus, including 4 types of NPC_EX, 5 types of NPC_IN; 7 types of EX, 5 types of EX_IN MIX and 2 types of APC/AST cells ( Fig S6B ). Notably, region-specific enrichment of neuronal subtypes was observed: NPC_EX2, NPC_IN16 (SVZ NPC), EX19 (SP neuron), EX5/EX11 (early-born L4), and APC/AST8 were gyrus-preferring, while NPC_IN7(SVZ NPC), EX26(SP neuron), EX2 (early-born L4), EX42 (late-born L2/3), and EX_IN_MIX2 were sulcus-enriched ( Figs 3D , S6B ). Subregional analysis revealed gyrus-to-sulcus gradients in cell-type proportions across all cortical layers (VZ, SVZ, IZ, SP, CP; Fig 3E , Table S2 ). APC/AST8 showed gyrus-biased distribution spanning VZ to IZ, while NPC_IN7 exhibited peak enrichment in SVZ and EX42 demonstrated maximal CP divergence with sulcus bias. Temporal quantification across developmental stages (GW11–GW21) revealed dynamic, region-specific shifts in NPCs (NPC_EX2,7,16; NPC_IN6,7,14,15), excitatory neurons (EX2,18,22,26,32,42,45), and glial/non-neuronal cells (APC/AST5,8; OPC/OL1,3; MIC; VLMC) ( Fig 3F ). To understand the contribution of regulatory factors to the asynchronous development of the sulci and gyri, the profile of differential gene expression in various sublayers between gyrus and sulcus were compared ( Figs S6C, D ; Table S3 ). A total of 48 DEGs were identified in the VZ, 50 in the SVZ, 50 in the IZ, 128 in the SP, and 194 in the CP ( Table S3 ). Among the upregulated DEGs in gyri, several genes were associated with neurogenesis and synapse organization, such as HOPX, PTN, VIM, and TUBB2A 37 , 38 , 39 , 40 , while others were linked to the functions of oligodendrocytes/astrocytes, including SPARC and GFAP 41 , 42 ( Fig S6C, S6D ). To elucidate the patterns of cell type-specific expression in the DEGs, we analyzed the profile of gene expression across distinct cortical cell types by using stereo-seq datasets from the gyrus and sulcus ( Fig 3G ). The analysis revealed that HOPX and SPARC were predominantly expressed in APC/AST and NPC_EX (including radial glia cells) ( Fig S6E ). However, HOPX exhibited significantly higher enrichment in gyrus-localized NPC_IN14 rather than any NPC_EX subclusters compared to its sulcus counterpart. PTN showed elevated expression in both NPC_IN14 and NPC_EX16, with gyrus-localized EX2 and APC/AST8 displaying higher PTN expression than their sulcus counterparts. Furthermore, FTL was highly expressed in NPC_IN4, NPC_EX16, EX19, EX2, and APC/AST8 within the gyrus ( Fig 3G ). These findings highlight region-specific molecular signatures across neuronal and glial cell populations during cortical development. The spatial distribution of DEGs related to cortical folding was analyzed in human fetal brains and smooth-brained species ( Figs 3H-K ). The gene expression patterns in the developing human fetal brain at GW15 were compared with those at equivalent developmental stages (E15.5-E16.5) of the mouse cortex 43 , 44 . In human GW15 cortex, HOPX was highly expressed in the VZ/SVZ, particularly in gyral VZ ( Fig 3H ), while mouse Hopx (E15.5-E16.5) showed uniform, sparse VZ distribution, in the Atlas of the Developing Mouse Brain ( MOSTA ) data 45 ( Fig 3I ). SPARC displayed broad human cortical expression but higher gyrus-specific subplate enrichment ( Fig 3J ), contrasting with its diffuse A-P gradient in mice ( Fig 3K ). Similarly, ID2 , ZEB1 , and VIM —human gyrus-sulcus markers— exhibited A-P gradients in E14.5/E16.5 mice ( Figs S6E-F ).The distinct but evolutionarily conserved spatial expression patterns of these genes in human and mouse cortices suggest their potential involvement in the development of cortical folding structures. The characteristics of D1/D2 hybrid MSNs in the developing striatum Medium spiny neurons (MSNs) were isolated from interneurons through sample library identifier-based segregation using datasets from both Shi et al. 15 and our independent experiments. snRNA-seq analysis of 18,024 cells revealed five transcriptionally distinct clusters classified as MSNs ( Figs 4A , S7A-B ). Notably, two clusters demonstrated concurrent expression of D1- and D2-type dopamine receptor markers, identifying a hybrid D1/D2 MSN subpopulation. Transcriptomic co-expression of prototypic markers TAC1 (D1-associated) and PENK (D2-associated) was validated across both snRNA-seq and ST datasets, further corroborating the existence of this hybrid MSN population ( Figs 4B , S4C ). Spatial transcriptomic analysis demonstrated significant spatial co-expression of TAC1 and PENK transcripts across striatal subdomains, indicative of a pronounced enrichment of D1-D2 co-expressing MSNs within striatal compartments relative to GE-derived neuronal populations ( Figs 4B , S4D ). Regional specificity was evident, with TAC1 exhibiting a marked enrichment in ventromedial striatal territories, whereas PENK expression was predominantly localized to dorsomedial striatal regions. Over time, the prevalence of D2-MSNs exhibits a progressive increase, whereas the proportion of D1/D2 hybrid populations demonstrates a concomitant decrease ( Fig S4D ). This dorsoventral stratification suggests divergent expression gradients of D1 and D2 MSN subtypes, further supporting functional and molecular heterogeneity within striatal circuits. These results provide direct evidence for the existence of a D1/D2 hybrid MSN subpopulation and highlight its anatomical predominance in the striatum. snRNA-seq analysis revealed distinct spatial enrichment patterns of D1/D2 hybrid MSN subtypes: D1/D2 hybrid MSN1 exhibited pronounced localization within the GE, while D1/D2 hybrid MSN2 demonstrated preferential enrichment in the striatum. During gestational maturation, the proportional abundance of D1/D2 hybrid MSN1 subtypes exhibited a progressive decline within both the GE and striatum, whereas the relative frequency of D1/D2 hybrid MSN2 subtypes showed a concomitant increase across both regions with advancing gestational age ( Figs 4C , S4E-F) . These region-specific distributions were further validated through integrative spatial mapping of snRNA-seq clusters onto ST datasets, which corroborated the anatomical segregation of D1/D2 hybrid MSN1 and D1/D2 hybrid MSN2 subtypes across the GE and striatum, respectively ( Figs 4D and 4E ) . Download figure Open in new tab Fig.S7 | Validation and comparative analysis of MSN subtypes, related to Fig.4 A. Jaccard similarity index quantifying overlap between MSN subtype classifications across independent clustering iterations. B. UMAP projection showing different data source and tissue regions. Left: UMAP projection integrating snRNA-seq data from this study and a reference dataset (Shi et al.). Right: UMAP stratification of MSNs by tissue origin (GE vs. striatum). C. UMAP visualization illustrating co-expression of TAC1 and PENK across MSN subtypes in snRNA-seq data. D. Stack bar plot showing the distribution of TAC1 and PENK as well as co-expression in different gestational weeks in striatum. E. Spatial transcriptomics colormap showing TAC1 and PENK co-expression within the GE. Scales: white = 1 mm, black = 100 μm; inset shows higher-magnification view. F. Bar plot comparing the proportional representation of MSN clusters across integrated snRNA-seq dataset. Download figure Open in new tab Fig. 4 | Spatiotemporal dynamics of D1/D2 hybrid MSNs. A. UMAP and dot plot showing clustering and specific markers of MSNs. Left: UMAP projection of MSNs isolated from integrated snRNA-seq data (see Fig. 1B ), partitioned into five transcriptionally distinct clusters. Right: Dot plot depicting cluster-specific marker expression. Two clusters were classified as D1/D2 hybrid MSNs (co-expressing canonical markers), two as D1_MSNs, and one as D2_MSNs. Bubble size shows the percentage of cell expressed; scale bar shows the average expression level. B. Spatial co-expression showing neuropeptide markers expression in striatal regions. Left: Representative spatial expression patterns of TAC1 and PENK . Right: Ridge plot quantifying TAC1 , PENK , and co-expression signal density along the ventral-dorsal axis. D: dorsal, L: lateral. Scales: white = 1 mm, black = 100 μm; inset shows higher-magnification view. C. Bar plot showing the proportional distribution of the five MSN clusters within the GE and striatum at distinct gestational weeks using snRNA data. GW: samples from this study; gw: published data by Shi et al. D. Integration of snRNA-seq and spatial transcriptomics data showing cell type distribution patterns in GE and STR. D: dorsal, L: lateral. Scales: 1mm. E. Stack-bar plot summarizing the regional and gestational stage-specific distribution of spatial cell types across ST sections in GE and striatum. See also Fig. S7 . The formation of patch and matrix compartments in the striatum The striatum is heterogeneously consisted of patch and matrix compartments, which are distinguished by their distinct neurochemicals 46 . To explore the spatial expression pattern in the developing striatum, we performed polynomial regression on the normalized gene expression data at each time point, resulting in 46 upregulated and 46 downregulated genes ( Fig S8A ). PCP4, PCDH17 and PENK showed higher expression on dorsal-lateral side ( Fig S8B ). To explore the dorsal-lateral versus ventral-medial genes of striatum, spatial variable gene (SVG) analysis was performed on the LGE-mantle zone (promising striatum) on coronal sections. Transcriptional profiling revealed distinct spatial enrichment patterns, with 4 genes exhibiting significant enrichment in the ventromedial region, while 25 genes showed pronounced enrichment in the dorsolateral region, suggesting region-specific functional specialization ( Fig S8C ) . Some genes (eg. NRGN, NTM) maintained region preference across all time points, whereas other genes ( GUCY1A1, CRABP1 and SIX3 ) did not ( Fig S8C ). At GW21, patch and matrix-like compartments were observed ( Figs 5A and S8D ). Notably, GABRA5 and EPHA4 , the known patch and matrix markers in the adult human brain 47 , colocalized with patch-enriched and matrix-enriched SVGs, respectively ( Fig S8E ). The patch and matrix marker genes exhibited compensatory expression, highlighting the unique features of these compartments in the striatum. Download figure Open in new tab Fig.S8 | Spatial transcriptomic identification of patch and matrix compartments, related to Fig.5 . A. Polynomial regression showing the up- or down-expression trend of genes across discrete gestational time points. The genes were z-score normalized and scaled. B. Spatial gene expression plot of significant genes selected from Fig S8A . Scales:1mm. C. Spatial expression of dorsal and ventral enriched genes at GW11, GW15, GW21. D: dorsal; L: lateral. Scales:1mm. D. Spatial expression of patch and matrix enriched genes at GW11, GW15, GW21. D: dorsal; L: lateral. Scales:1mm. E. Colocalization of novel and well-known patch/matrix genes. D: dorsal; L: lateral. Scales:1mm. F. Gaussian blur identifying regions of patch and matrix by SVGs at GW21. D: dorsal; L: lateral. Scales:1mm. Download figure Open in new tab Fig. 5 | Development of striatal patch and matrix compartments A. Spatial visualization of the spatial variable genes showing the structure of striatal patch and matrix at GW11, GW15, and GW21. D: dorsal; L: lateral. Scales: 1mm. B. Volcano plot showing DEGs of patch and matrix at GW21. C. Spatial visualization of patch and matrix gene module expression at GW11, GW15 and GW21. D: dorsal; L: lateral. Scales: 1mm. D. Venn plot showing the DEGs of patch and matrix shared by this GW21 spatial data and published adult human striatum data. E. The expression of SST and NPY in ISH fetal striatum at 15pcw and 21pcw from Allen Atlas. Scales:1mm. F. Spatial-ID showing spatially resolved single cell types of patch and matrix. Identified region of patch and matrix by Gaussian blur (upper left). Spatial visualization of enriched cells types in patch and matrix (lower left). Magnification of mapped cells (right). White dotted lines are the borders of patch and matrix. Red dotted lines are zoom-in zones. D: dorsal; L: lateral. Scales: 1mm. G. Stack-bar plot showing the percentage of cell types in patch and matrix. See also Figs. S8 , S9 , S10 and Table S4. Download figure Open in new tab Fig.S9 | Spatial visualization of top 40 patch DEGs, related to Fig.5 . A. Spatial visualization of top 40 patch DEGs. D: dorsal; L: lateral. Scales:1mm. Download figure Open in new tab Fig.S10 | Spatial analysis of patch and matrix DEGs, related to Fig.5 . A. Gaussian blur identifying regions of patch and matrix by SVGs at GW11 and GW15. D: dorsal; L: lateral. Scales:1mm. B. The fraction of cells expressing patch marker genes at GW11, GW15 and GW21. Each dot represents a patch DEG calculated. The red and blue lines show two representative high expression genes in patch. C. Spatial visualization of neuropeptide genes NPY, CENPS-CORT and CORT at GW11, GW15 and GW21. D: dorsal; L: lateral. Scales: 1mm. D. Expression of patch and matrix enriched top genes in ISH data of mouse brain at E11.5, E13.5, E15.5, E18.5, P4, P14 and P28 from Allen Brain Atlas. Scales: 1mm. E. Dot plot showing the expression of SST, NPY and CORT in striatum interneurons from snRNA data. To further analyze the developmental features of the patch and matrix compartments, we applied Gaussian blur to extract patch and matrix ( Fig S8F ; see STAR Methods ), resulting in 114 patch markers and 617 matrix marker genes ( Fig 5B ; Table S4 ). Pathways related to the response to thyroid hormone stimulus, synapse organization, innervation and regionalization were enriched in the patch. Conversely, in the matrix, the enriched GO terms were primarily related to adhesion molecules, axon regeneration and neuronal projection, indicating cell migration and axon development ( Table S4 ). These results suggested that distinct biological processes occur in the patch and matrix compartments. Gene module scores for both patch and matrix spatial gene sets based on DEGs were calculated across all time points ( Fig 5C ). The patch module demonstrated ubiquitous distribution throughout the dorsolateral striatum by GW 11 and progressively organized into distinct patch compartments at GW15 and GW21. In contrast, the matrix module exhibited a complementary spatial distribution pattern to the patch module by GW21, consistent with established striatal compartmentalization principles ( Fig 5C ). To investigate developmental-specific features as well as conserved genes in patch and matrix compartments, we compared patch and matrix markers with published data from the adult brain 47 ( Fig 5D ). Remarkably, CDH family was enriched in the adult patch and matrix but not in the fetal brain. Only 5 patch genes and 9 matrix genes were consistently related to the maintenance of the patch or matrix compartments. A total of 109 genes, such as SST , were enriched in developmental patches ( Fig 5D ). The temporal features of top 40 developmental patch genes were plotted ( Fig S9A ). Importantly, at GW15, SST exhibited a patch-like pattern, while other genes, such as SYNDIG1L, did not ( Figs 5A , S8D , S9A ). These findings suggest that the genes in patches gradually and asynchronously developed. Furthermore, SST was expressed in 98.1% of the identified patch cells at GW11 and 100% at GW15. Although the percentage decreased to 71.4% at GW21, it still remained above the average level of all genes ( Figs S10A and S10B ). Moreover, the expression patterns of CORT and CENPS-CORT ( both encoding cortistatin structurally similar to somatostatin), as well as NPY, were all similar to SST at GW15 and GW21 ( Fig S10C ). The expression of SST and NPY exhibited a patch-like pattern in ISH data of fetal striatum at 15pcw and 21pcw ( Fig 5E ). These findings indicated that SST is a patch-enriched gene in the human fetal brain, and these patch and matrix region-specific genes dynamically contribute to and maintain the properties of the developing striatum. We then explored patch and matrix genes in mouse embryo ( Fig S10D ) . Gabra5 and Gpr88 exhibited patched pattern at E18.5, Alcam at P4, however, Sst and Npy did not show patch enriched characteristics ( Fig S10D ). These suggested that despite the structures are conserved, the genes in the developing patches are different between human and mice. Subsequently, the spatial distribution of single cell types in patch and matrix were quantified ( Figs 5F , 5G ). At GW21, the proportion of D1/D2 hybrid MSN2 and NPC_IN11 were higher in patch, while D1_MSN1, NPC_IN10 and IN12 were greater in matrix ( Figs 5F , 5G ). Though at low levels, SST/NPY/CORT were co-expressed in the NPC_IN11 cells ( Fig S10E ). These results suggested neuropeptide SST and NPY may involve in patch formation. Overall, our results demonstrated that the patch and matrix undergo dynamic development. Discussion This study established a single-cell resolution spatial map of developing human brain from GW8 to GW21. By systematically analyzing the spatiotemporal gene expression and cellular distribution patterns, we reveal new mechanisms governing cortical expansion and uncover previously unrecognized developmental principles orchestrating striatal organization within the subpallium. Cortical expansion and arealization The developing cortex is anatomically categorized as the VZ, SVZ, IZ, SP and CP. SP was observed as a distinguished layer at GW10 after the emergence of the CP, which occurred at approximately GW8-GW9 3 . In addition to the concept of SP based on immunostaining, this study identified the subplate and its cellular composition at the transcriptional level ( Figs 2A , S5A ). The SP was separated from the preplate SP/CP after GW12. SP neurons in the SP layer are the earliest generated and mature neurons in the cerebral cortex, guiding neuronal migration towards the CP and establishing the earliest synapses 48 , 49 . Several studies have characterized human transcriptional features of SP-specific neurons in the human fetal brain 20 , 50 , 51 .The neurons (EX26/19/45/10 and EX_IN_MIX2) largely constituted the SP exhibited strong correlation with adult L5/6 neurons, align with findings that SP neurons differentiate into deep layer cortical neurons 52 , suggested a correlation of SP with layer 6 in adult brain. The cells in the subplate are dynamically changed during cortical expansion as more interneurons migrating to the cortical plate 32 . To elucidate the development of subplate and spatiotemporal segregation mechanisms between the subplate and layer 6, systematic analysis of fetal cortical samples across sequential developmental stages is required. Coinciding with the clear anatomical separation of SP and CP, deep and upper layer neurons underwent progressive laminar segregation concurrent with the emergence of distinct cortical lobes at GW21 ( Figs 1C , 2B ). Previous studies have identified that NPY is predominantly expressed in excitatory lineage cells within the parietal, temporal, and V1 cortices compared to the PFC, motor cortex, and somatosensory cortex 53 . However, these studies did not specify the neuronal subtypes expressing NPY and their laminar distribution, or their potential connectivity with other neuronal populations. Here, we identified a temporal lobe-enriched EX subtype, EX33, characterized by enriched NPY expression, residing in layer 2/3. This subtype exhibits potential functional interactions with L4 neuron EX2 through NPY-NPY1R ligand-receptor pairs ( Figs 2F-I , S5G ). The NPY plays critical roles in neural systems, such as promoting proliferation of postnatal neuronal precursor cells 54 , regulating anxiety and stress related responses 55 , feeding 56 , learning and memory 57 , 58 , endocrine function 59 , and circadian rhythms 60 . The function of NPY in L2/3 EX neurons remain to be addressed. In adult macaque V1 cortex, a type of EX neuron expressing NPY was also found restricted to L2/3 61 . Yet, the role of the NPY -expressing EX in cortical arealization remain understudied. In the developing mouse cortex, Npy expression becomes detectable post-E18.5, with higher expression levels observed in anterior cortical regions compared to posterior regions until postnatal day 4 (P4). By P14, Npy expression becomes uniformly distributed along the anterior-posterior axis, persisting through P28. ( https://developingmouse.brain-map.org/ ). The spatiotemporal expression pattern of Npy in the developing murine cortex is divergent from that observed in the human fetal cortex. These findings highlight the need for stratified analyses of NPY ’s functional roles across species and cortical layers. Transcriptomic features of developing fissures Recent studies have shown that specific molecules or modeled mechanicals promote the cortical folding in gyrencephalic species 62 , 63 , 64 , 65 . This study provided transcriptome resources for the patterns of gene expression in various developing fissures using the human fetal brain stereo-chip ( Figs 3A , S6A ). It was the first systematic identification of DEGs and specific celltypes from the gyrus and its adjacent sulcus in the human fetal cortex in situ. In human cortical folding, multiple sulci (e.g., Superior Frontal Sulcus, Callosal sulcus, Cingulate Sulcus, Sylvian fissure) and adjacent gyri were included for analysis ( Fig S6A ). While gene expression differences between gyri and sulci were subtle ( Fig S6C ), key divergences were observed in NPC/APC/EX ratios—particularly in SVZ and SP regions ( Figs 3D , 3E , Table S2 ). The gyrus exhibited significantly more HOPX -enriched APC/AST8 cells ( Fig 3G ), likely due to: (1) higher APC/AST8 and NPC_IN14 proportions (major factor), and (2) elevated HOPX expression in gyral NPC_IN14 (secondary factor). The characteristics of D1/D2 hybrid MSNs in the developing striatum D1/D2 hybrid MSNs have been documented in multiple species, including primates (e.g., macaques), rodents, and other model systems, across numerous studies 66 , 67 , 68 , 69 , 70 However, their presence during human neurodevelopment remains unconfirmed. In this study, we provide the first evidence of D1/D2 hybrid MSNs in the developing human brain, by identifying these cells across distinct developmental stages with striatal enrichment observed from GW15 onward. We further classified two subtypes of D1_D2 co-expressing MSNs: D1/D2 hybrid MSN1 exhibited preferential localization to the GE, whereas D1/D2 hybrid MSN2 demonstrated pronounced enrichment in the striatum. While a comparative analysis of these distinct clusters falls outside the scope of this investigation, their phenotypic and functional characterization represents a critical avenue for future research. Transcriptomic and cellular dynamics in the formation of patch and matrix compartments A previous study has demonstrated that striatal neurons in patch and matrix compartments undergo distinct developmental processes 9 . Comparative analysis with prior studies of the adult human striatum 47 revealed evolutionarily conserved genes enriched in striatal patch (e.g., GABRA5 ) and matrix (e.g., EPHA4 ) compartments. Additionally, we identified novel developmental-stage-specific markers ( ALCAM, GPR88, and SST ), which to our knowledge represent the first report of their dynamic expression patterns during striatal ontogeny ( Fig 5D ). Notably, SST was well colocalized within all patch compartments and was enriched in patches as early as GW11 ( Fig 5A ). Spatial profile of striatal cell subtypes showed compartment-specific enrichment, with NPC_IN11 interneurons (co-expressing SST , NPY , and CORT ) exhibiting a significantly higher prevalence within the patch compartment compared to the matrix. Moreover, the NPY showed a patch like pattern at GW15 and pcw21 from ISH data. However, the SST and NPY didn’t show a patch enriched pattern in developing mouse brain, and there was no difference in the density of SST between the patch and matrix in the adult mouse striatum 71 . These data implicate SST and NPY in human regulatory mechanisms underlying striatal compartmentalization, contrasting with murine models where these neuropeptides lack analogous roles. The spatiotemporal dynamics of patch-matrix developmental remodeling and the contributions of SST/NPY signaling to this process remain poorly characterized, necessitating systematic interrogation across postnatal-to-adult transitions. In summary, our study depicted a spatial-temporal map of developing regions in the brain by stereo-seq and snRNA-seq. We demonstrated dynamic structural changes associated with cortical layer expansion, gyrification, and subpallial striatal compartment development. This study provides novel insights into the development of the cortex and striatum and enhance our overall understanding of brain development. Limitations of the study The efficient capture of all genes during the experimental process was challenged by using Stereo-seq, particularly for those genes with low expression levels. This may lead to the inadvertent exclusion of significant genes from both processes of experimental and data analysis. Cells in close proximity may be erroneously binned together to increase statistical power of analysis. Due to the scarcity of human fetal brain samples, the obtainment of multiple replicates was not feasible for this study. Further experimentation is necessary to test and confirm the conclusions and hypotheses generated from this study. Methods HUMAN SUBJECTS Human embryo tissues were obtained in accordance from West China 2 nd University Hospital with ethic approval of Ethics Committees of West China 2 nd University Hospital (19H1194). All women gave written consent to the use of fetal tissues according to the ISSCR guidelines for fetal tissue donation. Fetal brain samples were collected after the donor patients signed an informed consent document that was in strict observance of the legal and institutional ethical regulations for elective pregnancy termination specimens at West China 2 nd University Hospital. The gestational age was measured in weeks from the first day of the woman’s last menstrual period. Tissue collection Fetal brains were collected in ice-cold artificial cerebrospinal fluid (ACSF) containing 125.0 mM NaCl, 26.0 mM NaHCO 3 , 2.5 mM KCl, 2.0 mM CaCl 2 , 1.0 mM MgCl 2 , 1.25 mM NaH 2 PO 4 at a pH of 7.4 when oxygenated (95% O 2 and 5% CO 2 ). The brain hemispheres were collected from three fetuses (11 gestational weeks, GW11; 15 gestational weeks, GW15; 16 gestational weeks, GW16; 21 gestational weeks, GW21). The isolated brain hemispheres were quickly wiped dry with sterile gauze and mixed thoroughly with 4°C optimal cutting temperature (OCT) compound (4583#, Sakura). Subsequently, the brain blocks were transferred to a metal mold fully embedded with 4°C OCT, quickly frozen on dry ice, and then stored in -80°C refrigerator. To minimize RNA degradation, all solutions were prepared with sterilized water containing diethyl pyrocarbonate (DEPC), and all instruments were washed with sterilized water containing DEPC and RNase Zap (AM9780, Invitrogen). The whole tissue collection process was completed within 30 minutes. Sample collection The 1/3, 2/5 and 1/2 coronal sections of GW11 and GW15 fetal brain, and 1/2 coronal section of GW21 fetal brain were collected for stereo-seq analysis. We aligned 4 chips to form a square and the 2/5 coronal section of GW15 was attached on the top. The 1/2 coronal section of GW15 was attached on the top of 3 aligned chips, and the 1/2 coronal section of GW21 was attached on the top of 14 aligned chips ( Figs 1C , S2C ). Each chip was then used for stereo-seq analysis. The adjacent tissue of each coronal section, except 1/3 section of GW15, was collected for snRNA analysis. In addition, the following tissues were also collected for snRNA analysis: the cortex of 1/3 coronal section of GW11, 1/3 coronal section of GW16, 1/2 coronal section of GW15 as well as four functional groups of cortex of 1/2 coronal section of GW21; the GE region on 1/2 coronal section of GW11 and GW15, and LGE and MGE region on 1/2 coronal section of GW21; the striatum on 1/2 coronal section of GW11, GW15 and GW21 respectively. Tissue-section preparation for snRNA-seq The DNBelab C Series Single-Cell Library Prep Set (MGI, 1000021082) was used as previously described 72 . Briefly, single-nucleus suspensions were used for droplet generation, emulsion breakage, bead collection, reverse transcription and cDNA amplification to generate barcoded libraries. Indexed libraries were constructed according to the manufacturer’s protocol. The sequencing libraries were quantified by Qubit ssDNA Assay Kit (Thermo Fisher Scientific). The sequencing libraries were sequenced by the DIPSEQ T7 sequencer at BGI-Shenzhen with the following sequencing strategy: 41-bp read length for read 1, and 100-bp read length for read 2. Tissue-section preprocessing for Stereo-seq Stereo-seq library preparation and sequencing were performed as previously described 73 . Briefly, tissues were adhered to the Stereo-seq chip. A nucleic acid dye staining (Thermo Fisher, Q10212) was performed to assess tissue morphology and quality. Tissue sections were permeabilized at 37° for 12 min and reverse transcription was conducted by second strand synthesis and cDNA denaturation. The cDNA was purified using AMPure XP beads (Vazyme, N411-03). The indexed single-cell RNA-seq libraries were constructed according to the manufacturer’s protocol. The sequencing libraries were then sequenced by the DIPSEQ T7 sequencer at BGI-Shenzhen with the following sequencing strategy: 50-bp read length for read 1, and 100-bp read length for read 2. snRNA data quality control and pre-processing The sequencing data were processed as previously described 11 . Nuclei with Unique Molecular Identifier counts (UMIs) counts <1500, or gene counts 20% were removed. Ribosomal genes, mitochondrial genes and blood-related genes ( HBA1, HBA2, HBB, HBG1, HBG2, HBQ1, HBD, HBM, HBE1 and HBZ ) were also removed according to published paper 74 . Data were log-normalized and top 3000 highly variable genes were selected using Seurat 75 LOESS method for downstream analysis. snRNA data integration for dimension reduction, clustering and subclustering The single-nuclei data from different sources 12 , 14 , 15 were integrated with data from this study, and the conservative gene sets of each single-nuclei data set were used for downstream analysis. The batch effects were corrected using harmony algorithm, and the data dimension was reduced and projected to UMAP. Cells were divided into three major categories based on the marker genes of the main excitatory neurons, inhibitory neurons and non-neuronal neurons. Subclustering was performed on each major cell group. Wilcoxon rank sum test was applied using the “sc.tl.rangk_genes_groups”function for differential gene expression analysis among clusters. Genes with log2 fold-change > 1 and FDR < 0.05 were retained as significant marker genes for the subcluster. Jaccard similarity was calculated with the top 50 ranked marker genes of each pair of cell clusters, and clusters were combined with Jaccard similarity larger than 0.8. 0.1% of the cells were identified as outliers and removed based on the cell distance to the center point. Subclustering analysis identified 136 distinct subtypes, which were subsequently aggregated into 12 biologically defined cell types based on their marker gene expression profiles. MILO analysis For snRNA, after data integration, the cells were then clustered to three trimesters and the differential abundant analysis was performed using python MILO package (v0.1.1). For ST data, after cell mapping, the mapped cell types in ST data were quantified and the same analysis was performed. Cross-method correlation analysis Gene specificity correlation analysis was used to evaluate the concordance between SpaGCN and FuseMap annotations. A unified set of 2,000 highly variable genes (HVGs) was identified from the integrated SpaGCN-FuseMap dataset using FindVariableFeatures function (Seurat; selection method = ‘vst’, mean-variance threshold = 1.5). Gene specificity vectors between homologous subregions (e.g., VZ_FuseMap vs. VZ_SpaGCN) were compared using rank-based Spearman correlation. Statistical significance was determined through two-tailed permutation testing (10,000 iterations) with Benjamini-Hochberg correction for multiple hypothesis testing. Gyrus-Sulcus subregions Pairwise Cluster Correlations The stereo-seq data of gyrus/sulcus regions were lassoed by their anatomically spatial structure, including Superior frontal sulcus (SFS), Callosal sulcus (CaS), Cingulate sulcus (CiS), Sylvian fissure (SF) 76 , 77 , provisional prigeminal fissure (PPF) 78 , and their adjacent gyri. For gyrus/sulcus profiling, we manually lassoed three paired gyral and sulcal regions from anatomically defined 1/3 anterior, 2/5 middle, and 1/2 posterior coronal sections of GW15 brains. Two gyral and three sulcal regions were segmented from GW21 mid-posterior coronal sections, with technical replicates (n=4 or 5 sections per region) ( Figs S6A ). The region of interest was selected on ssDNA (using Adobe Photoshop v25.3.1). Then the expression matrix was aligned to ssDNA and the region of interest was lassoed. The subregions (VZ/SVZ/IZ/SP/CP) were annotated by FuseMap. The subregion gene expression was normalized using the SCTransform function in Seurat. The correlation analysis between gyrus and sulcus consisted of:1) the selection of a common gene set for the comparison. Variable genes identified from the gyrus and sulcus combined data using Seurat FindVariableFeatures function (selection.method = ‘vst’, nfeatures = 2,000) were used to generated averaged scaled expression values and correlated. 2) the calculation of Spearman rank correlations (and their p-values) of gene specificities among all pairs of subregions in the gyrus and sulcus. Comparative analysis with published atlas Reference subregional definitions 12 were reprocessed using FuseMap-based harmonization. Transcriptomically matched subregions were identified through reciprocal nearest neighbor alignment (integration weight = 0.7) in PCA-UMAP latent space (dims=1:30). Subsequent correlation analyses followed the aforementioned Spearman framework. Transfer fetal brain celltype to adult brain celltype XGBoost (eXtreme Gradient Boosting) model 79 was trained to transfer our fetal brain cell type annotation to adult brain single-cell atlas. The model objective was set as multi-classification task (evaluation metrics: mlogloss, learning rate=0.001). The model was trained on 1 Nvidia A100-80G GPU. Sanky plot The ST data at same gestational week were combined, and the data from adjacent gestational weeks were used to calculate Jaccard similarity and z-score. R package networkD3 (0.4) was then used to generate Sanky Plot. Differential expression genes (DEGs) and GO enrichment analysis Seurat FindAllMarkers function was used to calculate DEGs with minimum expression percentage as 0.25. The other parameters were set as default. Calculated DEGs with adjusted p values > 0.5 were filtered out, and the remaining DEGs were ordered by descending average log 2 FC value. ClusterProfiler 80 package and org.Hs.eg.db 81 package were used to calculate enriched GO categories based on the NIH database tool DAVID (v6.7) ( https://david.ncifcrf.gov ) 82 , 83 . Ligand-receptor communication analysis CellChat 84 package was used to calculate the ligand-receptor communication of mapped cells. Spatial data (ST data) pre-processing Expression profile of matrix of human brain was divided into non-overlapping bins covering an area of 100*100 DNB to form bin100 data (∼50 µm) and the transcripts of the same gene were aggregated within each bin. For GW15 and GW21 chips, the chips were aligned according to chip coordinates respectively. Cluster ST data using SpaGCN Spatial data of bin100 were clustered using SpaGCN 17 for each coronal section. The resolution and region identification were set according to Brainspan atlas of the developing human brain ( https://www.brainspan.org/static/atlas ). SpaGCN was employed to identify spatial domains jointly by using a graph convolutional neural network to learn a representation aggregating gene expression data from surrounding spots to identify the domains. The adjacency graph used in the convolution was constructed based on spatial location. The resolution and region identification were set according to Brainspan atlas of the developing human brain ( https://www.brainspan.org/static/atlas ). Lasso the region of interest in ST data The region of interest was selected on ssDNA. Then the expression matrix was aligned to ssDNA and the region of interest was lassoed. Spatial Transcriptomics cell segmentation The optimal cell segmentation model (cell_segmentation_v3.0.onnx) developed by Stereopy 85 was applied for Stereo-seq data with extremely dense cell populations to perform nuclear segmentation on human fetal brain nucleic acid staining images. Subsequently, the segmented cell masks were integrated with Stereo-seq data using Stereopy’s stereo.tl.read_gef function, generating spatial transcriptomic data with single-cell resolution for human fetal brain tissues. The entire pipeline can be accessed at https://stereopy.readthedocs.io/en/latest/Tutorials/Cell_Segmentation.html Spatial Transcriptomics clustering by FuseMap at single cell resolution The FuseMap 81 model was used to cluster ST data. Each spatial transcriptomics sample’s unique characteristics were modeled using dedicated autoencoders and graph autoencoders, forming a sample-specific probabilistic generative model. During the model training process, 15% of the cells were hold out as the validation set. Training was designed to early stop if there was no improvement in the validation loss for consecutive epochs. The training procedure included a pretraining phase, a weighted adversarial alignment as employed in previous research, and a final training phase that utilized cell-specific weights to ensure effective integration for cases when cell type compositions vary intrinsically. To further identify finer brain regions, Leiden 86 clustering was applied to the spatial embedding of all spatial transcriptomics samples. Joint clustering was performed using the function scanpy.tl.leiden at a resolution 1. Mapping of snRNA data to ST data using SpatialID We employed Spatial-ID 87 , an effective and reliable cell type annotation algorithm, to annotate cells within Stereo-seq data with the cell types clustered according to snRNA analyses to exhibit the spatial distribution of various cell types. Initially, we calculated the intersecting set of genes present in both the snRNA and Stereo-seq data. These genes were then used as input features during the training of a Deep Neural Network (DNN) model. Leveraging class-weight Cross Entropy Loss, the DNN model was trained over 30 epochs at a learning rate of 3e-4. This proficiently trained DNN model harbored the mapping information from gene profiles to predefined cell types. It was subsequently applied to cells in the Stereo-seq data to establish the initial probabilities of different cell types. Following this, the spatial neighborhood details of the Stereo-seq cells were encapsulated into an adjacency matrix. This matrix was weighted by normalized Euclidean distances. We then utilized a Graph Convolution Network (GCN) composed of two auto-encoders and a classifier block. This network integrated the gene profiles and spatial neighborhood information of the cells in spatial transcriptomic data to generate the final probabilities of cell types. The two auto-encoders received gene profiles and the adjacency matrix respectively, to create latent cell representations. Meanwhile, the classifier block established a connection from these representations to the initial cell type probabilities predicted by the DNN model. Training of these elements was achieved through self-supervised and classification losses using an adversarial learning strategy over a course of 200 epochs. The cell type that showed the highest value in the final probabilities was designated as the type of a cell. Spatial-ID utilized in this study was implemented using PyTorch (v1.13.1) and PyTorch Geometric (v2.1.0) in Python (v3.8.10). Cortical cell distribution analysis and visualization For each ST data, the cell number of different cell types that mapped to different layers using SpatialID was quantified, and R package scatterpie (v0.1.8) was used to generate pie/ stacked bar chart showing the distribution of different cell types in each layer in the whole cortex or gyrus /sulcus groups. Differential abundance (gyrus vs. sulcus) was tested by Fisher’s exact test (FDR <0.1) and plotted as bubble maps (ggplot2 v3.4.2; size= -log10[FDR]). HotSpot gene module calculation To identify gene enrichment modules in spatial transcriptomics data, a spatial adjacency matrix was constructed using positional data. Then, HotSpot 33 was used to detect distinct gene modules. Finally, the significant gene module was obtained by applying an FDR threshold of <0.05. Gene module visualization Gene module scores were calculated using Seurat AddModuleScore function, and visualized in ST data using SpatialPlot function. SpaTrack infer cell trajectory in ST data SpaTrack 88 was used to infer cell trajectory in ST data (n_neigh_pos = 100). The trajectory was then generated by using matplotlib.pyplot.streamplot function in Python. Ridge plot of genes distribution pattern The tissue was rotated to make x axis align with dorsal-ventral axis. Then the cells expressed TAC1 and/or PENK were quantified along x axis, and the density ridge plot was generated respectively. Calculate spatial variable genes (SVGs) The region of interest was extracted using Spateo package as previously described. The spatial variable genes were calculated using Spateo package Moran I algorithm. The genes were filtered with Moran I score > 0.15 and p_val<0.05. Tope SVGs were plotted respectively. Gaussian blur and label transfer to extract patch and matrix Selected SVGs were used to calculate the expression level on ST data. Gaussian blur was then used to extract patch and matrix structure from ST data. The annotation results from bin100 data were transferred to cellbin data. Specifically, a square region (side length: 100 arbitrary spatial units) was defined around the centroid coordinates of each bin100 point. All cellbin coordinates falling within this region were programmatically assigned the annotation corresponding to the central bin100 point, thereby establishing a spatially resolved annotation mapping between the two coordinate systems. Venn Diagram DEGs for patch and matrix compartment were identified using the rank_gene_groups function in Scanpy. Genes exhibiting a log2 fold change (logFC) exceeding 0.5 were then intersected with those previously reported for patch and matrix regions in the referenced study 47 , respectively, and the overlaps were visualized using Venn diagrams. Polynomial regression The Spateo Moran’s I algorithm was employed to identify spatially variable genes (SVGs) within individual ST section datasets. Significant genes were subsequently filtered by retaining those for which Moran’s I values in any three consecutive gestational weeks exceeded their gene-specific mean Moran’s I values computed across all gestational weeks. The polynomial regression analysis was then used to model the expression trends (ascending or descending) of genes across discrete gestational time points, based on z-score normalized and scaled gene expression profiles. Quantification and statistical analysis This study did not use statistical methods to determine the sample size. The experiments were not randomized and the investigators were not blinded to the allocation during both the experiments and outcome assessment. To assess region-specific differences in cellular abundance between gyrus and sulcus compartments, Chi-square tests were performed on integer-normalized cell type proportions (counts converted to whole-number percentages) with the chi2_contingency function in SciPy (v1.11.0). This normalization accounted for disparities in total cell counts across anatomical regions. For stereo-seq data of the gyrus and sulcus, following subregion annotation, differential expression analysis was performed using Wilcoxon Rank tests followed by Bonferroni correction for multiple tests. Statistical significance was set at p < 0.05, except for stereo-seq data which was set at padj < 0.01. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this paper, the authors used Copilot on STOmics Cloud ( https://cloud.stomics.tech ) to polish language. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. Data availability The raw sequence data have been deposited to CNGB Nucleotide Sequence Archive, and temporary access links can be requested from the lead contact. The data that support the findings of this study have been deposited into CNGB Sequence Archive (CNSA) 89 of China National GeneBank DataBase (CNGBdb) 90 with accession number CNP0005272 ( http://db.cngb.org/cnsa/project/CNP0005272_8c0dc42f/reviewlink/ ). This paper does not provide the original code, the programs and packages used in the analysis are listed in the key resources table. Further details necessary for reanalyzing the data presented in this paper can be obtained upon request from the lead contact. Author contributions L.Z, X.L and Yaoyao.Z designed the project; Yunjia.Z, Y.L, Xunan.S, Xiaobo.S, T.Z, Z.W, Z.Z, L.C, S.Q, J.Y, and F.H performed analysis; Investigation, Xue.X, C.S, C.L, M.L, F.W, Yaqi.L and J.L performed experiments and discussion; Yanxin.L, J.J., Xue.X, and Yaoyao.Z collected the samples; Yunjia.Z, Y.L, Xunan.S, and C.S wrote the draft,; Yunjia.Z, Y.L, Xunan.S, C.S, Q.Z, Z.L, S.Z, Z.D, and L.L reviewed and edited the manuscript; Yunjia.Z, L.Z, S.L, and Yaoyao.Z supervised the project; X.X, S.L, L.Z, and Yaoyao.Z provide funding support. Competing interests The authors declare no competing interests. Additional information Correspondence and requests for materials should be addressed to Yaoyao Zhang. supplemental Excel table titles and legends Table S1. Hotspot gene module, related to Figure 2 Table S2. Cell proportions of gyri and sulci in each sublayer, related to Figure 3 Table S3. DGEs of gyri and sulci in each sublayer, related to Figure 3 Table S4. DEGs and GO pathways of patch and matrix at GW21, related to Figure 5 Acknowledgments This work was supported by Sichuan Science and Technology Program (2024YFFK0072); National Key Research and Development Program of China (2022YFC3400400 to X.X.); National Key R&D Program of China (2021YFA0805100); the Frontiers Medical Center, Tianfu Jincheng Laboratory Foundation (TFJCPI20250038) and the STI 2030-Major Projects (2021ZD0200500). Project analysis was performed on the STOmics Cloud ( https://cloud.stomics.tech ). This paper is from the Mesoscopic Brain Mapping Consortium. Funder Information Declared Sichuan Science and Technology Program , 2024YFFK0072 National Key Research and Development Program of China , 2022YFC3400400 to X.X. National Key R&D Program of China , 2021YFA0805100 Frontiers Medical Center, Tianfu Jincheng Laboratory Foundation STI 2030-Major Projects , 2021ZD0200500 Footnotes ↵ 17 Lead contact References 1. ↵ Ouyang M , Dubois J , Yu Q , Mukherjee P , Huang H . Delineation of early brain development from fetuses to infants with diffusion MRI and beyond . Neuroimage 185 , 836 – 850 ( 2019 ). OpenUrl CrossRef PubMed 2. ↵ Westbrook A , Frank MJ , Cools R . A mosaic of cost–benefit control over cortico-striatal circuitry . Trends in Cognitive Sciences 25 , 710 – 721 ( 2021 ). OpenUrl CrossRef PubMed 3. ↵ Bystron I , Blakemore C , Rakic P . Development of the human cerebral cortex: Boulder Committee revisited . Nature Reviews Neuroscience 9 , 110 – 122 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 4. ↵ Llinares-Benadero C , Borrell V . Deconstructing cortical folding: genetic, cellular and mechanical determinants . Nature Reviews Neuroscience 20 , 161 – 176 ( 2019 ). OpenUrl CrossRef PubMed 5. ↵ Lewitus E , Kelava I , Kalinka AT , Tomancak P , Huttner WB . An adaptive threshold in mammalian neocortical evolution . PLoS biology 12 , e1002000 ( 2014 ). OpenUrl CrossRef PubMed 6. ↵ Albin RL , Young AB , Penney JB . The functional anatomy of basal ganglia disorders . Trends in neurosciences 12 , 366 – 375 ( 1989 ). OpenUrl CrossRef PubMed Web of Science 7. ↵ Zhou F-M. The striatal medium spiny neurons: what they are and how they link with Parkinson’s disease . In: Genetics, Neurology, Behavior, and Diet in Parkinson’s Disease ). Elsevier ( 2020 ). 8. ↵ Bocchi VD , et al. The coding and long noncoding single-cell atlas of the developing human fetal striatum . Science 372 , eabf5759 ( 2021 ). OpenUrl Abstract / FREE Full Text 9. ↵ van der Kooy D , Fishell G . Neuronal birthdate underlies the development of striatal compartments . Brain research 401 , 155 – 161 ( 1987 ). OpenUrl CrossRef PubMed Web of Science 10. ↵ Fishell G , van der Kooy D . Pattern formation in the striatum: developmental changes in the distribution of striatonigral projections . Developmental Brain Research 45 , 239 – 255 ( 1989 ). OpenUrl CrossRef PubMed 11. ↵ Liu C , et al. A portable and cost-effective microfluidic system for massively parallel single-cell transcriptome profiling . 818450 ( 2019 ). 12. ↵ Li Y , et al. Spatiotemporal transcriptome atlas reveals the regional specification of the developing human brain . Cell 186 , 5892 – 5909 . e5822 ( 2023 ). OpenUrl CrossRef PubMed 13. ↵ Qiu X , et al. Spateo: multidimensional spatiotemporal modeling of single-cell spatial transcriptomics . BioRxiv , 2022.2012. 2007. 519417 ( 2022 ). 14. ↵ Velmeshev D , et al. Single-cell analysis of prenatal and postnatal human cortical development . Science 382 , eadf0834 ( 2023 ). OpenUrl CrossRef PubMed 15. ↵ Shi Y , et al. Mouse and human share conserved transcriptional programs for interneuron development . Science 374 , eabj6641 ( 2021 ). OpenUrl CrossRef PubMed 16. ↵ He Y , et al. Towards a universal spatial molecular atlas of the mouse brain . bioRxiv , 2024.2005. 2027. 594872 ( 2024 ). 17. ↵ Hu J , et al. SpaGCN: Integrating gene expression, spatial location and histology to identify spatial domains and spatially variable genes by graph convolutional network . Nature methods 18 , 1342 – 1351 ( 2021 ). OpenUrl PubMed 18. ↵ Pollen AA , et al. Molecular identity of human outer radial glia during cortical development . Cell 163 , 55 – 67 ( 2015 ). OpenUrl CrossRef PubMed 19. ↵ Zhong S , et al. A single-cell RNA-seq survey of the developmental landscape of the human prefrontal cortex . Nature 555 , 524 – 528 ( 2018 ). OpenUrl CrossRef PubMed 20. ↵ Polioudakis D , et al. A single-cell transcriptomic atlas of human neocortical development during mid-gestation . Neuron 103 , 785 – 801 . e788 ( 2019 ). OpenUrl CrossRef PubMed 21. ↵ Zhong S , et al. Decoding the development of the human hippocampus . Nature 577 , 531 – 536 ( 2020 ). OpenUrl CrossRef PubMed 22. ↵ Viho EM , Buurstede JC , Berkhout JB , Mahfouz A , Meijer OC . Cell type specificity of glucocorticoid signaling in the adult mouse hippocampus . Journal of neuroendocrinology 34 , e13072 ( 2022 ). OpenUrl CrossRef PubMed 23. ↵ Flandin P , et al. Lhx6 and Lhx8 coordinately induce neuronal expression of Shh that controls the generation of interneuron progenitors . Neuron 70 , 939 – 950 ( 2011 ). OpenUrl CrossRef PubMed Web of Science 24. ↵ Tang K , Rubenstein JL , Tsai SY , Tsai M-J . COUP-TFII controls amygdala patterning by regulating neuropilin expression . Development 139 , 1630 – 1639 ( 2012 ). OpenUrl Abstract / FREE Full Text 25. ↵ García MT , et al. Transcriptional profiling of sequentially generated septal neuron fates . Elife 10 , e71545 ( 2021 ). OpenUrl CrossRef PubMed 26. ↵ Inoue T , Ota M , Ogawa M , Mikoshiba K , Aruga J . Zic1 and Zic3 regulate medial forebrain development through expansion of neuronal progenitors . Journal of Neuroscience 27 , 5461 – 5473 ( 2007 ). OpenUrl Abstract / FREE Full Text 27. ↵ Wang Q , et al. Organization of the connections between claustrum and cortex in the mouse . Journal of Comparative Neurology 525 , 1317 – 1346 ( 2017 ). OpenUrl CrossRef PubMed 28. ↵ Hoerder-Suabedissen A , et al. Temporal origin of mouse claustrum and development of its cortical projections . Cerebral Cortex 33 , 3944 – 3959 ( 2023 ). OpenUrl CrossRef PubMed 29. ↵ Lee M , et al. Tcf7l2 plays crucial roles in forebrain development through regulation of thalamic and habenular neuron identity and connectivity . Developmental Biology 424 , 62 – 76 ( 2017 ). OpenUrl CrossRef PubMed 30. ↵ Heide M , et al. Lhx5 controls mamillary differentiation in the developing hypothalamus of the mouse . Frontiers in neuroanatomy 9 , 113 ( 2015 ). 31. ↵ Lebow MA , Chen A . Overshadowed by the amygdala: the bed nucleus of the stria terminalis emerges as key to psychiatric disorders . Molecular psychiatry 21 , 450 – 463 ( 2016 ). OpenUrl CrossRef PubMed 32. ↵ Hoerder-Suabedissen A , Molnár Z . Development, evolution and pathology of neocortical subplate neurons . Nature Reviews Neuroscience 16 , 133 – 146 ( 2015 ). OpenUrl CrossRef PubMed 33. ↵ DeTomaso D , Yosef N . Hotspot identifies informative gene modules across modalities of single-cell genomics . Cell systems 12 , 446 – 456 . e449 ( 2021 ). OpenUrl CrossRef PubMed 34. ↵ Gerald C , et al. A receptor subtype involved in neuropeptide-Y-induced food intake . Nature 382 , 168 – 171 ( 1996 ). OpenUrl CrossRef PubMed 35. ↵ Golombek DA , Biello SM , Rendon RA , Harrington ME . Neuropeptide Y phase shifts the circadian clock in vitro via a Y2 receptor . Neuroreport 7 , 1315 – 1319 ( 1996 ). OpenUrl CrossRef PubMed Web of Science 36. ↵ Colmers WF , Klapstein GJ , Fournier A , St-Pierre S , Treherne KA . Presynaptic inhibition by neuropeptide Y in rat hippocampal slice in vitro is mediated by a Y2 receptor . British journal of pharmacology 102 , 41 ( 1991 ). 37. ↵ Berg DA , et al. A common embryonic origin of stem cells drives developmental and adult neurogenesis . Cell 177 , 654 – 668 . e615 ( 2019 ). OpenUrl CrossRef PubMed 38. ↵ Li H , et al. Pleiotrophin ameliorates age-induced adult hippocampal neurogenesis decline and cognitive dysfunction . Cell reports 42 , ( 2023 ). 39. ↵ Bittermann E , et al. Differential requirements of tubulin genes in mammalian forebrain development . PLoS genetics 15 , e1008243 ( 2019 ). OpenUrl 40. ↵ Widestrand Å , et al. Increased neurogenesis and astrogenesis from neural progenitor cells grafted in the hippocampus of GFAP−/− Vim−/− mice . Stem Cells 25 , 2619 – 2627 ( 2007 ). OpenUrl CrossRef PubMed Web of Science 41. ↵ Kucukdereli H , et al. Control of excitatory CNS synaptogenesis by astrocyte–secreted proteins Hevin and SPARC . Proceedings of the National Academy of Sciences 108 , E440 – E449 ( 2011 ). OpenUrl Abstract / FREE Full Text 42. ↵ Lee Y , Messing A , Su M , Brenner M . GFAP promoter elements required for region-specific and astrocyte–specific expression . Glia 56 , 481 – 493 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 43. ↵ Zhou Y , Song H , Ming G-l . Genetics of human brain development . Nature Reviews Genetics 25 , 26 – 45 ( 2024 ). OpenUrl CrossRef PubMed 44. ↵ Sepp M , et al. Cellular development and evolution of the mammalian cerebellum . Nature 625 , 788 – 796 ( 2024 ). OpenUrl CrossRef PubMed 45. ↵ Thompson CL , et al. A high-resolution spatiotemporal atlas of gene expression of the developing mouse brain . Neuron 83 , 309 – 323 ( 2014 ). OpenUrl CrossRef PubMed 46. ↵ Brimblecombe KR , Cragg SJ . The striosome and matrix compartments of the striatum: a path through the labyrinth from neurochemistry toward function . ACS chemical neuroscience 8 , 235 – 242 ( 2017 ). OpenUrl PubMed 47. ↵ Matsushima A , et al. Transcriptional vulnerabilities of striatal neurons in human and rodent models of Huntington’s disease . Nature communications 14 , 282 ( 2023 ). 48. ↵ Grant E , Hoerder-Suabedissen A , Molnár Z . Development of the corticothalamic projections . Frontiers in neuroscience 6 , 53 ( 2012 ). 49. ↵ Ohtaka-Maruyama C . Subplate neurons as an organizer of mammalian neocortical development . Frontiers in Neuroanatomy 14 , 8 ( 2020 ). 50. ↵ Miller JA , et al. Transcriptional landscape of the prenatal human brain . Nature 508 , 199 – 206 ( 2014 ). OpenUrl CrossRef PubMed Web of Science 51. ↵ Wang WZ , et al. Subplate in the developing cortex of mouse and human . Journal of anatomy 217 , 368 – 380 ( 2010 ). OpenUrl CrossRef PubMed Web of Science 52. ↵ Ozair MZ , Kirst C , van den Berg BL , Ruzo A , Rito T , Brivanlou AH . hPSC modeling reveals that fate selection of cortical deep projection neurons occurs in the subplate . Cell stem cell 23 , 60 – 73 . e66 ( 2018 ). OpenUrl CrossRef PubMed 53. ↵ Bhaduri A , et al. An atlas of cortical arealization identifies dynamic molecular signatures . Nature 598 , 200 – 204 ( 2021 ). OpenUrl CrossRef PubMed 54. ↵ Hansel D , Eipper B , Ronnett G . Neuropeptide Y functions as a neuroproliferative factor . Nature 410 , 940 – 944 ( 2001 ). OpenUrl CrossRef PubMed Web of Science 55. ↵ Johnson DA. Handbook of Neurochemistry and Molecular Neurobiology: Sensory Neurochemistry ( 2007 ). 56. ↵ Dodt S , et al. NPY-mediated synaptic plasticity in the extended amygdala prioritizes feeding during starvation . Nature communications 15 , 5439 ( 2024 ). OpenUrl PubMed 57. ↵ Chen T , et al. The Drosophila NPY-like system protects against chronic stress–induced learning deficit by preventing the disruption of autophagic flux . Proceedings of the National Academy of Sciences 120 , e2307632120 ( 2023 ). OpenUrl CrossRef PubMed 58. ↵ Gøtzsche C , Woldbye D . The role of NPY in learning and memory . Neuropeptides 55 , 79 – 89 ( 2016 ). OpenUrl CrossRef PubMed 59. ↵ McDonald JK , Koenig JI . Neuropeptide Y actions on reproductive and endocrine functions . In: The biology of neuropeptide Y and related peptides ). Springer ( 1993 ). 60. ↵ Yannielli P , Harrington M . Neuropeptide Y in the mammalian circadian system: effects on light-induced circadian responses . Peptides 22 , 547 – 556 ( 2001 ). OpenUrl CrossRef PubMed Web of Science 61. ↵ Wei J-R , et al. Identification of visual cortex cell types and species differences using single-cell RNA sequencing . Nature communications 13 , 6902 ( 2022 ). OpenUrl PubMed 62. ↵ Liu J , et al. The Primate-Specific Gene TMEM14B Marks Outer Radial Glia Cells and Promotes Cortical Expansion and Folding . Cell Stem Cell 21 , 635 – 649 .e638 ( 2017 ). OpenUrl CrossRef PubMed 63. ↵ Long KR , et al. Extracellular Matrix Components HAPLN1, Lumican, and Collagen I Cause Hyaluronic Acid-Dependent Folding of the Developing Human Neocortex . Neuron 99 , 702 – 719 .e706 ( 2018 ). OpenUrl CrossRef PubMed 64. ↵ Kawasaki H . Molecular Investigations of the Development and Diseases of Cerebral Cortex Folding using Gyrencephalic Mammal Ferrets . Biol Pharm Bull 41 , 1324 – 1329 ( 2018 ). OpenUrl PubMed 65. ↵ Karzbrun E , Kshirsagar A , Cohen SR , Hanna JH , Reiner O . Human Brain Organoids on a Chip Reveal the Physics of Folding . Nat Phys 14 , 515 – 522 ( 2018 ). OpenUrl CrossRef PubMed 66. ↵ Zillich L , et al. Cell type-specific Multi-Omics Analysis of Cocaine Use Disorder in the Human Caudate Nucleus . Res Sq , ( 2024 ). 67. ↵ Gokce O , et al. Cellular Taxonomy of the Mouse Striatum as Revealed by Single-Cell RNA-Seq . Cell Rep 16 , 1126 – 1137 ( 2016 ). OpenUrl CrossRef PubMed 68. ↵ Gayden J , et al. Integrative multi-dimensional characterization of striatal projection neuron heterogeneity in adult brain . bioRxiv , ( 2023 ). 69. ↵ Phan BN , et al. Single nuclei transcriptomics in human and non-human primate striatum in opioid use disorder . Nat Commun 15 , 878 ( 2024 ). 70. ↵ He J , et al. Transcriptional and anatomical diversity of medium spiny neurons in the primate striatum . Curr Biol 31 , 5473 – 5486 .e5476 ( 2021 ). OpenUrl CrossRef PubMed 71. ↵ Banghart MR , Neufeld SQ , Wong NC , Sabatini BL . Enkephalin Disinhibits Mu Opioid Receptor-Rich Striatal Patches via Delta Opioid Receptors . Neuron 88 , 1227 – 1239 ( 2015 ). OpenUrl CrossRef PubMed 72. ↵ Han L , et al. Cell transcriptomic atlas of the non-human primate Macaca fascicularis . 604 , 723 – 731 ( 2022 ). OpenUrl 73. ↵ Chen A , et al. Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball-patterned arrays . 185 , 1777 – 1792 . e1721 ( 2022 ). OpenUrl 74. ↵ Zhou X , et al. Deciphering the spatial-temporal transcriptional landscape of human hypothalamus development . 29 , 328 – 343 . e325 ( 2022 ). OpenUrl 75. ↵ Hao Y , et al. Integrated analysis of multimodal single-cell data . 184 , 3573 – 3587 . e3529 ( 2021 ). OpenUrl 76. ↵ Zhang Z , et al. Development of the fetal cerebral cortex in the second trimester: assessment with 7T postmortem MR imaging . AJNR Am J Neuroradiol 34 , 1462 – 1467 ( 2013 ). OpenUrl Abstract / FREE Full Text 77. ↵ Kerwin J , et al. The HUDSEN Atlas: a three-dimensional (3D) spatial framework for studying gene expression in the developing human brain . J Anat 217 , 289 – 299 ( 2010 ). OpenUrl CrossRef PubMed Web of Science 78. ↵ Proshchina A , et al. Neuromorphological Atlas of Human Prenatal Brain Development: White Paper . Life (Basel ) 13 , ( 2023 ). 79. ↵ Chen T , Guestrin C . XGBoost: A Scalable Tree Boosting System ( 2016 ). 80. ↵ Wu T , et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data . Innovation (Camb ) 2 , 100141 ( 2021 ). 81. ↵ Carlson M , Falcon S , Pages H , Li N. org. Hs. eg. db: Genome wide annotation for Human . R package version 3 , 3 ( 2019 ). 82. ↵ Huang da W , Sherman BT , Lempicki RA . Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources . Nat Protoc 4 , 44 – 57 ( 2009 ). OpenUrl CrossRef PubMed Web of Science 83. ↵ Huang da W , Sherman BT , Lempicki RA . Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists . Nucleic Acids Res 37 , 1 – 13 ( 2009 ). OpenUrl CrossRef PubMed Web of Science 84. ↵ Jin S , et al. Inference and analysis of cell-cell communication using CellChat . Nat Commun 12 , 1088 ( 2021 ). OpenUrl CrossRef PubMed 85. ↵ Fang S , et al. Stereopy: modeling comparative and spatiotemporal cellular heterogeneity via multi-sample spatial transcriptomics . Nat Commun 16 , 3741 ( 2025 ). OpenUrl CrossRef PubMed 86. ↵ Traag VA , Waltman L , Van Eck NJ . From Louvain to Leiden: guaranteeing well-connected communities . Scientific reports 9 , 1 – 12 ( 2019 ). OpenUrl CrossRef PubMed 87. ↵ Shen R , et al. Spatial-ID: a cell typing method for spatially resolved transcriptomics via transfer learning and spatial embedding . Nat Commun 13 , 7640 ( 2022 ). OpenUrl CrossRef PubMed 88. ↵ Shen X , et al. Inferring cell trajectories of spatial transcriptomics via optimal transport analysis . Cell systems 16 , 101194 ( 2025 ). 89. ↵ Guo X , et al. CNSA: a data repository for archiving omics data . Database (Oxford) 2020 , ( 2020 ). 90. ↵ Chen FZ , et al. CNGBdb: China National GeneBank DataBase . Yi Chuan 42 , 799 – 809 ( 2020 ). OpenUrl PubMed View the discussion thread. Back to top Previous Next Posted August 15, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum Yunjia Zhang , Youning Lin , Xunan Shen , Xue Xiao , Cai Song , Xiaobo Shi , Tao Zhou , Yanxin Li , Zihan Wu , Zhenkun Zhuang , Chunqiong Li , Meng Li , Feng Wen , Jianlin Liu , Qiangqiang Zhang , Zhao-Lu Li , Songbo Zhang , Lei Cao , Susu Qu , Yaqi Li , Jianhua Yao , Fubaoqian Huang , Xin Liu , Ziqing Deng , Longqi Liu , Xun Xu , Jianwei Jiao , Li Zhang , Shiping Liu , Yaoyao Zhang bioRxiv 2025.08.14.670213; doi: https://doi.org/10.1101/2025.08.14.670213 Share This Article: Copy Citation Tools Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum Yunjia Zhang , Youning Lin , Xunan Shen , Xue Xiao , Cai Song , Xiaobo Shi , Tao Zhou , Yanxin Li , Zihan Wu , Zhenkun Zhuang , Chunqiong Li , Meng Li , Feng Wen , Jianlin Liu , Qiangqiang Zhang , Zhao-Lu Li , Songbo Zhang , Lei Cao , Susu Qu , Yaqi Li , Jianhua Yao , Fubaoqian Huang , Xin Liu , Ziqing Deng , Longqi Liu , Xun Xu , Jianwei Jiao , Li Zhang , Shiping Liu , Yaoyao Zhang bioRxiv 2025.08.14.670213; doi: https://doi.org/10.1101/2025.08.14.670213 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Bioinformatics Subject Areas All Articles Animal Behavior and Cognition (7635) Biochemistry (17691) Bioengineering (13892) Bioinformatics (41936) Biophysics (21452) Cancer Biology (18588) Cell Biology (25504) Clinical Trials (138) Developmental Biology (13378) Ecology (19899) Epidemiology (2067) Evolutionary Biology (24320) Genetics (15609) Genomics (22506) Immunology (17736) Microbiology (40394) Molecular Biology (17181) Neuroscience (88605) Paleontology (666) Pathology (2832) Pharmacology and Toxicology (4824) Physiology (7641) Plant Biology (15153) Scientific Communication and Education (2045) Synthetic Biology (4294) Systems Biology (9825) Zoology (2271)