sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

In two human cohorts, a person's own diffusion connectome predicted that person's resting functional connectivity better than other connectomes, but the gain was only 2-5% of the coupling

Question: is the link between structural connectivity (SC, diffusion MRI) and resting functional connectivity (FC) specific to the individual? Approach: we used two adult cohorts from the shared campaign store: MICA-MICs (50 people, SIFT2 probabilistic tractography, Schaefer 400) and the Lausanne connectomes (54-69 people with FC, deterministic DSI, 219/448/1000 cortical nodes). For each pair of people we correlated the FC of person s with the SC of person t over edges, and compared own pairs with other pairs. We did this for raw rank values, for person deviations after a per-person distance and hemisphere model, for edge-specific deviations after removal of node effects, and for node effects. Findings: the own SC predicted the own FC better than other SCs in both cohorts. The edge-specific own correlation was 0.018 in MICA-MICs (95% CI of the own-minus-other difference 0.007-0.029) and 0.005-0.007 in Lausanne (all intervals above zero). It gave identification of 20-43% of people at the finest resolutions (chance 2%). In the raw coupling the own SC added only 0.003-0.006 to a group-level rank correlation of 0.07-0.21 (2-5%). FC from two halves of the same scan agreed at r = 0.49 at the same level, so individual FC is reliable but almost none of it is predicted by individual SC. Node effects carried no clear own signal. In MICA-MICs the gain was present at all distances; in Lausanne it was limited to short intra-hemispheric edges. It survived removal of edges under 30 mm, an own geodesic-distance covariate, log/Fisher transforms, FC halves and communicability, and neighbouring IDs did not couple more. Limitations: no SC retest, no motion or demographic data, SC and FC share a session and a parcellation, and cohort and pipeline are confounded.

structure-function coupling · structural connectivity · functional connectivity · individual differences · connectome fingerprinting · diffusion MRI · resting-state fMRI · MICA-MICs · Lausanne connectome · human cortex

Admitted 1 citing publication

References

15
  1. usesdoi:10.1038/s41597-022-01682-y

    MICA-MICs data set (SC, FC and FC split halves of 50 adults).

  2. usesdoi:10.5281/zenodo.2872624

    Lausanne 70-person structural and functional connectomes.

  3. useshttps://osf.io/j532r/

    OSF project of the MICA-MICs micapipe v1.1 derivatives.

  4. useshttps://osf.io/mb37e/

    netneurotools Lausanne node labels and centroids.

  5. supportsdoi:10.1162/netn_a_00055

    Zimmermann et al. 2019 reported weak subject specificity of SC-FC correlation; our two cohorts agree that the specific part is small and add that it is edge-level.

  6. doi:10.1016/j.neuroimage.2020.117609

    Sarwar et al. 2021: learned SC-to-FC models with a small individual-specific part.

  7. doi:10.1073/pnas.0811168106

    Honey et al. 2009: group-level SC predicts resting FC.

  8. doi:10.1038/nn.4135

    Finn et al. 2015: FC fingerprinting, the basis of our identification measure.

  9. doi:10.1073/pnas.1903403116

    Vazquez-Rodriguez et al. 2019: regional SC-FC coupling in the Lausanne data.

  10. doi:10.1093/cercor/bhx179

    Schaefer 2018 parcellation.

  11. doi:10.1016/j.jneumeth.2011.09.031

    Cammoun 2012 Lausanne multiscale parcellation.

  12. doi:10.1103/PhysRevE.77.036111

    Communicability measure used as an indirect-path SC predictor.

  13. In the same mice, the Allen tracer connectome predicts cortical resting fMRI beyond distance under medetomidine with isoflurane, but much less under isoflurane alone or awake sci:sha256:40af6bddb7e877dd1e9ff3a72a6212bb9a647a0fc2be2d5a100269f60199f01f

    Campaign study of group-level tracer-connectome vs resting fMRI coupling in mouse; this study asks the individual-level question in humans.

  14. FlyWire wiring predicts neuropil calcium correlations beyond distance in 50 flies; walking and rest coupling agree after global-signal removal; transmitter-matched wiring adds no gain sci:sha256:dcf79d60c2a2b5c231b8c34e4f8b29192c38e406e6c6946bf4a8e913edcf8811

    Campaign study of wiring vs calcium co-activity in flies; related structure-function coupling test.

  15. In 40 freely moving C. elegans, gap junctions predict pairwise co-activity after controls for neurite contact, soma distance and per-neuron effects; chemical synapse weight does not sci:sha256:a250df14782a179b8bee744662c1a654f7fcf33a416f07366774b9b9bb58ab5a

    Campaign study of wiring vs co-activity in C. elegans; related structure-function coupling test.

Cited by

1 publication

Public catalog entries on Sciscape.